digital mapping techniques ‘99— workshop proceedingsthe digital mapping techniques...

U.S. Department of the InteriorU.S. Geological Survey

Digital Mapping Techniques ‘99—Workshop Proceedings

Edited by David R. Soller

May 19–22, 1999Madison, Wisconsin

Convened by theAssociation of American State Geologistsand theUnited States Geological Survey

Hosted by theWisconsin Geological and Natural History Survey

U.S. GEOLOGICAL SURVEY OPEN-FILE REPORT 99–3861999

This report is preliminary and has not been reviewed for conformity with U.S. Geological Survey editorial standards.

Any use of trade, product, or firm names in this publication is for descriptive purposes only and does not imply endorsement by the

U.S. Government or State governments.

III

CONTENTS

IntroductionBy David R. Soller (U.S. Geological Survey). . . . . . . . . . . . . . . . . . . . . . . 1

Oral PresentationsNational Atlas of the United States of America

By Gregory J. Allord (U.S. Geological Survey). . . . . . . . . . . . . . . . . . . 5

Geologic Mapping at the Oklahoma Geological Survey:The Move From Traditional to Digital CartographyBy T. Wayne Furr (Oklahoma Geological Survey) . . . . . . . . . . . . . . . . 7

Digital Geologic Map Production and Database Developmentin the Central Publications Group of the Geologic Division,U.S. Geological SurveyBy Diane E. Lane, Alex Donatich, F. Craig Brunstein, and

Nancy A. Shock (U.S. Geological Survey) . . . . . . . . . . . . . . . . . . . 11

Avenza’s MAPublisherBy Susan Muleme (Avenza Software Marketing Inc.) . . . . . . . . . . . . . 17

DesignJet Large Format Printer Trends and TechnologiesBy Alberto Berry (Hewlett-Packard Company) . . . . . . . . . . . . . . . . . . 19

“Can’t See the Geology for the Ground Clutter” — Shortcomings of the Modern Digital Topographic BaseBy David J. McCraw (New Mexico Bureau of Mines and

Mineral Resources). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21

The Evolution of Topographic Mapping in theU.S. Geological Survey’s National Mapping ProgramBy Robert M. Lemen (U.S. Geological Survey) . . . . . . . . . . . . . . . . . . 27

The National Geologic Map Database — A progress reportBy David R. Soller (U.S. Geological Survey) andThomas M. Berg (Ohio Geological Survey) . . . . . . . . . . . . . . . . . . . . 31

Proposed Guidelines for Inclusion of Digital Map Productsin the National Geologic Map DatabaseBy David R. Soller (U.S. Geological Survey),

Ian Duncan (Virginia Division of Mineral Resources),Gene Ellis (U.S. Geological Survey),Jim Giglierano (Iowa Geological Survey),and Ron Hess (Nevada Bureau of Mines and Geology) . . . . . . . . . . 35

Digital Map Production and Publication by Geological SurveyOrganizations; A Proposal for Authorship and Citation GuidelinesBy C. Rick Berquist, Jr. (Virginia Division of Mineral Resources) . . . 39

Plain-Language Resources for Metadata Creators and ReviewersBy Peter N. Schweitzer (U.S. Geological Survey) . . . . . . . . . . . . . . . . 43

ArcInfo 8: New GIS technology from ESRIBy Mike Price (Environmental Systems Research Institute, Inc.). . . . . 47

Extracting the Information from 3D Geologic Models —Maps and Other Useful OutputsBy Skip Pack (Dynamic Graphics, Inc.) . . . . . . . . . . . . . . . . . . . . . . . 49

Ground Water Atlas of the United StatesBy Gary D. Latzke (U.S. Geological Survey) . . . . . . . . . . . . . . . . . . . 51

Acquisition of Geologic Quadrangle Data from aDigital Index Map—A New ArcView Applicationto Geologic Data QueryBy Xin-Yue Yang, Warren H. Anderson, Thomas N. Sparks,and Daniel I. Carey (Kentucky Geological Survey) . . . . . . . . . . . . . . . 53

Progress Toward Development of a Standard Geologic Map Data ModelBy the Geologic Map Data Model Steering Committee . . . . . . . . . . . . 57

Geologic Concept Modeling, with Examples for Lithologyand some other Basic Geoscience FeaturesBy Stephen M. Richard (Arizona Geological Survey) . . . . . . . . . . . . . 59

Data Model for Single Geologic Maps: An Application of theNational Geologic Map Data ModelBy Donald L. Gautier (U.S. Geological Survey). . . . . . . . . . . . . . . . . . 77

Using the Proposed U.S. National Digital Geologic MapData Model as the Basis for a Web-Based GeoscienceLibrary PrototypeBy Boyan Brodaric, J.M. Journeay, and

S.K. Talwar (Geological Survey of Canada) . . . . . . . . . . . . . . . . . . 83

The Geometry of a Geologic Map DatabaseBy Ronald R. Wahl (U.S. Geological Survey) . . . . . . . . . . . . . . . . . . . 93

Geomatter: A Map-Oriented Software Tool forAttributing Geologic Map Information According to theProposed U.S. National Digital Geologic Map Data ModelBy Boyan Brodaric, Eric Boisvert, and

Kathleen Lauziere (Geological Survey of Canada) . . . . . . . . . . . . . 101

Development of the Kansas Geologic Names Database:A Link to Implementing the Geologic Map Data Model StandardsBy David R. Collins and Kurt Look (Kansas Geological Survey) . . . . 107

From Functional Analysis to CD—Digital Compilation of theTimmins Map Sheet, Abitibi Greenstone Belt, OntarioBy N.F. Trowell, J.A. Ayer, R. Berdusco, Z. Madon,

S. van Haaften, and A. Yeung (Ontario Geological Survey) . . . . . . 113

CONTENTS

V

Integration of Digital Geologic Databases and a Spatial MapDatabase in KentuckyBy Warren H. Anderson, Thomas N. Sparks, Jason Patton,

and Xin-Yue Yang (Kentucky Geological Survey) . . . . . . . . . . . . . 123

Geologic Mapping and Collection of Geologic Structure Datawith a GPS Receiver and a Personal Digital Assistance (PDA)ComputerBy Gregory J. Walsh, James E. Reddy, and

Thomas R. Armstrong (U.S. Geological Survey) . . . . . . . . . . . . . . 127

ArcView, a Geologic Mapping ToolBy Harold W. Baker (Missouri Geological Survey Program) . . . . . . . 133

Digital Resource Database for Management Decisions inCity of Rocks National ReserveBy David R. Bedford and

David M. Miller (U.S. Geological Survey) . . . . . . . . . . . . . . . . . . . 139

Present and Future Issues of a Mapping ProgramBy Frank Ganley (Alaska Division of Geological

and Geophysical Surveys) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145

Alabama Aquifer Vulnerability CD-ROM Project:Digital Publication of GIS-Derived Map ProductsBy Berry H. Tew and

Ruth T. Collier (Geological Survey of Alabama) . . . . . . . . . . . . . . 147

Geologic Resources Inventory for the National Park System:Status, Applications, and Geology-GIS Data ModelBy Joe Gregson, Anne Poole, Steve Fryer, Bruce Heise,

Tim Connors, and Kay Dudek (National Park Service). . . . . . . . . . 151

Poster SessionsExamples of Map Production from the Alaska Division

of Geological and Geophysical SurveysBy Andrew M. McCarthy and Frank Ganley (Alaska Division

of Geological & Geophysical Surveys) . . . . . . . . . . . . . . . . . . . . . 163

Geologic Map of the Monterey 1:100,000 Scale Quadrangle andMonterey Bay: A Digital DatabaseBy David L. Wagner (California Division of Mines

and Geology), H. Gary Greene (Moss Landing MarineLaboratory), Jason D. Little, Sarah E. Watkins, andCynthia L. Pridmore (California Division of Minesand Geology) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165

Geological Map Production at the Geological Survey of CanadaBy Vic Dohar and Dave Everett (Geological Survey of Canada). . . . . 167

A Digital Geologic Map Data Model Designed for GIS UsersBy Loudon R. Stanford, Kurt L. Othberg, Reed S. Lewis,

and Roy M. Breckenridge (Idaho Geological Survey) . . . . . . . . . . 169

VCONTENTS

Digital Geologic Map of the Harrodsburg 30 x 60-MinuteQuadrangle, Central KentuckyBy Warren H. Anderson and Thomas N. Sparks

(Kentucky Geological Survey) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171

Different Needs, Different Bases: Struggling with TopographicBases for Digital Geologic MapsBy Larry N. Smith and Susan M. Smith

(Montana Bureau of Mines and Geology) . . . . . . . . . . . . . . . . . . . 173

Evolution of Digital Geologic Mapping at the North DakotaGeological SurveyBy Karen J.R. Mitchell, Ann M.K. Fritz, and

Ryan Waldkirch (North Dakota Geological Survey) . . . . . . . . . . . . 177

The Integration of Sinkhole Data and Geographic InformationSystems: An Application Using ArcView 3.1By William E. Kochanov (Pennsylvania Geological Survey) . . . . . . . 183

The Making of the 1:250,000-Scale Digital Geologic Mapof PennsylvaniaBy Christine E. Miles and Thomas G. Whitfield

(Pennsylvania Geological Survey) . . . . . . . . . . . . . . . . . . . . . . . . . 185

The Development of a Digital Surface and SubsurfaceGeologic - Map Database of Charleston County, South CarolinaBy Peter G. Chirico (U.S. Geological Survey) . . . . . . . . . . . . . . . . . . 187

Using Regions and Route Systems to Model CompoundObjects in Arc/InfoBy Michele E. McRae (U.S. Geological Survey) . . . . . . . . . . . . . . . . 195

Production of USGS Map I-2669 — A folio of maps and 3-D imagesBy David R. Soller, Susan D. Price, and

D. Paul Mathieux (U.S. Geological Survey) . . . . . . . . . . . . . . . . . . 197

Production of USGS Map I-1970 — A Regional Map ShowingThickness and Character of Quaternary SedimentsBy David R. Soller and Will Stettner (U.S. Geological Survey) . . . . . 199

Offset Printing of Raster Image FilesBy Elizabeth V.M. Campbell (Virginia Division of Mineral Resources)

and Robert B. Fraser (U.S. Geological Survey) . . . . . . . . . . . . . . . 201

Quaternary Landforms in Wisconsin — How Hillshading CanBe Used To Accentuate Topographic FeaturesBy Chip Hankley (Wisconsin Geological and

Natural History Survey). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205

Appendix A. List of Attendees at the Digital MappingTechniques ‘99 Workshop. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211

Appendix B. Workshop Web Site. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213

Appendix C. List of Addresses, Telephone Numbers, and URLsfor Software and Hardware Suppliers . . . . . . . . . . . . . . . . . . . . . . . . . 215

CONTENTS

The Digital Mapping Techniques ‘99 (DMT’99)workshop was attended by 91 technical experts from 42agencies, universities, and private companies, includingrepresentatives from 30 state geological surveys (seeAppendix A). This workshop was similar in nature to thefirst two meetings, held in June, 1997, in Lawrence,Kansas (Soller, 1997), and in May, 1998, in Champaign,Illinois (Soller, 1998a). This year’s meeting was hosted bythe Wisconsin Geological and Natural History Survey,from May 19 to 22, 1999, on the University of Wisconsincampus in Madison. As in the previous meetings, theobjective was to foster informal discussion and exchangeof technical information. When, based on discussions atthe workshop, an attendee adopts or modifies a newlylearned technique, the workshop clearly has met thatobjective. Evidence of learning and cooperation amongparticipating agencies continued to be a highlight of theDMT workshops (see example in Soller, 1998b, and vari-ous papers in this volume).

The meeting’s general goal was to help move the stategeological surveys and the USGS toward development ofmore cost-effective, flexible, and useful systems for digitalmapping and geographic information systems (GIS) analy-sis. Through oral and poster presentations and special dis-cussion sessions, emphasis was given to: 1) methods forcreating and publishing map products (here, “publishing”includes Web-based release); 2) continued development ofthe National Geologic Map Database; and 3) progresstoward building a standard geologic map data model.Especially to support the interest in map preparation andpublication, five representatives of the GIS hardware andsoftware vendor community were invited to participate.

The three annual DMT workshops were coordinatedby the AASG/USGS Data Capture Working Group, whichwas formed in August, 1996, to support the Association ofAmerican State Geologists and the USGS in their effort tobuild a National Geologic Map Database (see Soller andBerg, this volume, and <http://ncgmp.usgs.gov/ngmdbpro-

ject/standards/datacapt/datacaptureWG.html>). TheWorking Group was formed because increased productionefficiencies, standardization, and quality of digital mapproducts were needed to help the Database, and the Stateand Federal geological surveys, provide more high-qualitydigital maps to the public.

ACKNOWLEDGMENTS

I thank the Wisconsin Geological and Natural HistorySurvey (WGNHS), and their Chief and State Geologist,James Robertson, for hosting this very productive andenjoyable meeting. I especially thank Mindy James(WGNHS), who coordinated the meeting, provided excel-lent support for the attendees, and maintained the meet-ing’s web site (see Appendix B). Her expertise and senseof humor are greatly appreciated. Thanks also to MikeCzechanski, Chip Hankley, Rilla Hinkes, MarciaJesperson, Deb Patterson, Kathy Roushar, and KathieZwettler (all WGNHS) for helping with the meeting logis-tics, Phil O’Leary (U. Wisconsin) for allowing us to usethe meeting facilities, and Dave Carlson (USGS) for set-ting up the online registration. I also note with gratitudethe contributions of the following individuals: Tom Berg(Chair, AASG Digital Geologic Mapping Committee) forhis help in conducting the meeting and for his continuedsupport of AASG/USGS efforts to collaborate on theNational Geologic Map Database; the members of theData Capture Working Group (Warren Anderson,Kentucky Geological Survey; Rick Berquist and ElizabethCampbell, Virginia Division of Mines and Geology; RobKrumm and Barb Stiff, Illinois State Geological Survey;Scott McColloch, West Virginia Geological and EconomicSurvey; Gina Ross, Kansas Geological Survey; DaveWagner, California Division of Mines and Geology; andTom Whitfield, Pennsylvania Geological Survey) foradvice in planning the workshop’s content and the sugges-

1

Introduction

By David R. Soller

U.S. Geological Survey908 National CenterReston, VA 20192

Telephone: (703) 648-6907Fax: (703) 648-6937

e-mail: [email protected]

tions to authors; and Adam Davis (USGS) for help withAppendix C. Finally, I thank all attendees for their partici-pation; their enthusiasm and expertise were the primaryreasons for the meeting’s success.

PRESENTATIONS

The workshop included 32 oral presentations. Nearlyall are supported by a short paper contained in theseProceedings. Some presentations were coordinated withDiscussion Sessions, described below. The papers repre-sent approaches that currently meet some or all needs fordigital mapping at the respective agency. There is not, ofcourse, a single “solution” or approach to digital mappingthat will work for each agency or for each program orgroup within an agency — personnel and funding levels,and the schedule, data format, and manner in which wemust deliver our information to the public require thateach agency design their own approach. However, thevalue of this workshop, and other forums like it, is throughtheir role in helping to design or refine these agency-spe-cific approaches to digital mapping and to find approachesused by other agencies that are applicable. In other words,communication helps us to avoid “reinventing the wheel.”

Most presentations ranged across a number of issues,so I make little attempt to organize the papers by topic.With my apologies to authors whose work I may not ade-quately describe, I provide here a brief description of eachpaper. For the sake of brevity, the lead or presentingauthor only is listed. Further information about the soft-ware and hardware referred to below and elsewhere inthese Proceedings is provided in Appendix C.

1. Gregory J. Allord (U.S. Geological Survey)—develop-ment of the new National Atlas of the United States ofAmerica, a government-wide effort to deliver mapinformation to the public.

2. T. Wayne Furr (Oklahoma Geological Survey)—evolu-tion of digital cartographic methods at the OklahomaGeological Survey.

3. Todd Fitzgibbon (U.S. Geological Survey)—anapproach for creating, reviewing, and releasing digitalgeologic map products (paper not supplied).

4. Diane E. Lane (U.S. Geological Survey)—methods fordigital geologic map production and database develop-ment in the Central Publications Group.

5. Susan Muleme (Avenza Software Marketing Inc.)—introduction to MaPublisher plugin for AdobeIllustrator.

6. Alberto Berry (Hewlett-Packard Company )—currentand future technologies for large format DesignJetplotters.

7. David McCraw (New Mexico Bureau of Mines andMineral Resources)—challenges to integrating digitaltopographic bases with geologic maps.

8. Robert Lemen (U.S. Geological Survey)—creation ofbase map products in the USGS National MappingProgram.

9. David R. Soller (U.S. Geological Survey)—progressreport on the National Geologic Map Database.

10. David R. Soller (U.S. Geological Survey)—specialdiscussion session: proposed guidelines for inclusionof digital map products in the National Geologic MapDatabase.

11. Rick Berquist (Virginia Division of MineralResources)—authorship and citation of digital geolog-ic maps and spatial data.

12. Peter Schweitzer (U.S. Geological Survey)—plain-language resources for metadata creators and review-ers.

13. Mike Price (Environmental Systems ResearchInstitute, Inc.)—introduction to ArcInfo 8: New GIStechnology from ESRI.

14. Skip Pack (Dynamic Graphics, Inc.)—introduction toEarthVision: extracting information from 3D geologicmodels.

15. Gary D. Latzke (U.S. Geological Survey)—productionmethods and products in the series Ground WaterAtlas of the United States.

16. Xin-Yue Yang (Kentucky Geological Survey )—anArcView tool for automating selection of map datafrom a geologic map database.

17. Data Model Steering Committee—special discussionsession: development of a draft standard geologic mapdata model and the Steering Committee.

18. Stephen M. Richard (Arizona Geological Survey)—data model concepts, and thoughts on implementingparts of the draft standard geologic map data model.

19. Donald L. Gautier (U.S. Geological Survey)—apply-ing the draft standard data model to single geologicmap products.

20. Boyan Brodaric (Geological Survey of Canada)—using the draft standard data model as the basis for aweb-based geoscience library prototype.

21. Ronald R. Wahl (U.S. Geological Survey)—problemsin representing spatial objects with typical GIS soft-ware.

22. Eric Boisvert (Geological Survey of Canada)—a soft-ware tool for attributing geologic maps in the draftstandard data model format.

DIGITAL MAPPING TECHNIQUES ‘992

23. David R. Collins (Kansas Geological Survey)—devel-opment of the Kansas Geologic Names Database, andpossible links to the draft standard data model.

24. Brian Berdusco (Ontario Geological Survey)—func-tional analysis of GIS, development of digital map-ping methods, and application to compilation of ageologic map.

25. Warren H. Anderson (Kentucky Geological Survey)—creating a statewide digital geologic map database.

26. Gregory J. Walsh (U.S. Geological Survey)—usingGPS and hand-held computers for geologic mapping.

27. Harold W. Baker (Missouri Geological Survey)—applying ArcView to geologic map compilation andproduction.

28. David R. Bedford (U.S. Geological Survey)—creatinga geologic map and digital photolibrary for manage-ment applications.

29. Frank Ganley (Alaska Division of Geological andGeophysical Surveys)—review of agency mappingprogram, and update of progress in implementing dig-ital mapping.

30. Nick Tew (Geological Survey of Alabama)—digitalgeologic map production, and geologic map applica-tions to groundwater vulnerability studies.

31. Steve Fryer (National Park Service)—the geologicresources inventory program.

32. Jeffrey M. Hyatt (Intergraph Corp.)—overview of car-tography and GIS products from Intergraph (paper notsupplied).

POSTERS

More than 15 posters were exhibited throughout theworkshop. These posters provided an excellent focus fortechnical discussions and support for oral presentations.Most are documented with a paper in these Proceedings,following the oral presentations.

DISCUSSION SESSIONS

To provide the opportunity to consider a topic in somedetail, special discussion sessions were held. Theseaddressed: 1) proposed guidelines for inclusion of digitalmap products in the National Geologic Map Database; 2)progress toward development of a draft standard geologicmap data model; and 3) a general discussion of ideas pre-sented during the meeting. Discussion session #1 led torevisions to the draft proposal (Soller, Duncan, Ellis,Giglierano, and Hess, this volume) and the recommenda-tion to submit the guidelines to management for considera-tion. Session #2 provided increased understanding of thenew process being used to promote development of thestandard data model, and session #3 provided recommen-dations for new features to add to future DMT meetings.

THE NEXT DMT WORKSHOP

At discussion session #3, it was decided that a fourthannual DMT meeting would be held, next year. Whileplanning for that event, the Data Capture Working Groupwill carefully consider the recommendations offered byDMT’99 attendees.

REFERENCES

Soller, D.R., editor, 1997, Proceedings of a workshop on digitalmapping techniques: Methods for geologic map data cap-ture, management, and publication: U.S. Geological SurveyOpen-File Report 97-269, 120 p,<http://ncgmp.usgs.gov/pubs/of97-269/>.

Soller, D.R., editor, 1998a, Digital Mapping Techniques ‘98—Workshop Proceedings: U.S. Geological Survey Open-FileReport 98-487, 134 p, <http://pubs.usgs.gov/openfile/of98-487/>.

Soller, D.R., 1998b, Introduction, in: D.R. Soller, ed., DigitalMapping Techniques ‘98—Workshop Proceedings: U.S.Geological Survey Open-File Report 98-487, p. 1-3,<http://pubs.usgs.gov/openfile/of98-487/intro.html>.

INTRODUCTION 3

5

A new National Atlas was authorized in 1997 toupdate the original Atlas published in 1970. The firstNational Atlas was generally found in reference collectionsof libraries as well as being used by educators and govern-ment organizations. This edition was a 400 page, 12pound printed book with 765 maps that took 7 years toproduce. The $100 price was a deterrent for purchase forhome use. This new National Atlas is intended to reachaudiences not typically addressed by U. S. Governmentprograms and products. Initial Atlas design efforts con-centrated on identifying customers, determining theirexpectations, and using this market research to refine prod-uct definitions. The new National Atlas is designed forindividuals who own powerful home computers.

The National Atlas of the United States of America™(http://www.nationalatlas.gov) will include four distinct

products: high-quality small-scale maps; authoritative,documented national geospatial and geostatistical datasets; easy-to-use web-based software for data display,query and custom information; and hot links to other siteson the WWW for up-to-date, real-time, and regional data.The U.S. Geological Survey, the lead agency in thisGovernment-wide effort, is also developing partnerships tomake a product that is responsive to the needs of sec-ondary markets like education, businesses, and libraries.The National Atlas will realize these goals for delivery ofinformation through the WWW and on CD-ROM productsthrough a combination of Government support as well asCooperative Research and Development Agreementssigned with one or more private businesses.

The National Atlas of the United States of America

By Gregory J. Allord

U.S. Geological Survey505 Science Drive

Madison, WI 53711-1061Telephone: (608) 238-9333 x200

Fax: (608) 238-9334e-mail: [email protected]

7

HISTORY

The Oklahoma Geological Survey (OGS) has main-tained a strong commitment to mapping the geology andnatural resources of the State of Oklahoma; a commitmentthat began before statehood with its predecessor organiza-tion, the Oklahoma Territorial Geological Survey. In 1904,before becoming the first State Geologist, Dr. Charles N.Gould, prepared a preliminary geologic map of theOklahoma Territory. Three years later, Gould contributedto the development of the enabling act for the creation ofthe state geological survey. When comparing length ofservice to other state geological surveys OGS may be con-sidered an infant. However, OGS claims the distinction ofbeing the only state geological survey in the nation to have beencreated under a directive of the constitution of a newly formed state.

The objectives and duties of the new survey weredefined by Oklahoma’s First Legislature. Senate Bill No.75 provided for a study of the geological formations of theState with special reference to its mineral deposits, includ-ing the preparation and publication of reports with maps.The reports are to provide both general and detaileddescriptions of the State’s geological resources.

The first full-color geologic map of Oklahoma, com-piled by Hugh D. Miser of the U.S. Geological Survey(USGS), was released in 1926. After 20 years of use, themap not only was outdated it was out of print as well. In1947, Miser returned to Norman to supervise revision ofthe 1926 map. Miser and 10 additional authors are givencredit for the geologic map of Oklahoma that was releasedin 1954 (Ham, 1983, p. 7). Forty-five years later the mapis still in use, but in need of major revisions. Mapping andother kinds of field studies have continued with the com-pletion of 145 Bulletins, 100 Circulars, 35 MineralReports, 31 Guidebooks, 5 Educational Publications, 35Geologic Maps, 9 Hydrologic Atlases, 65 SpecialPublications, and 45 Open-File Reports.

MAPPING PROGRAMS

Currently, OGS is conducting four kinds of geologicalmapping programs: county, resource, 7.5-minute quadran-gle, and 1:100,000 digital compilation. The purpose ofthese programs is to provide a better understanding of theState’s geological history and resources. The knowledgegained will enable public policy-makers and industry tosafely and wisely utilize Oklahoma’s geological resources.

For more than 70 years, county mapping has been thecornerstone of geological investigations at OGS. KayCounty, located in north-central Oklahoma is the mostrecent to be mapped. Currently, geologic investigationsare underway in the western part of Osage County.Traditionally, investigations in the county geological map-ping program have been produced in the Bulletin andCircular Series with maps at a scale of 1:63,360.

Recent resource mapping in Oklahoma has focused onthe coal reserves and environmental problems associatedwith past mining, including poor reclamation techniques.The principal user of the OGS coal-resource maps isindustry. Several mines in the State have been developedbased on information provided from these studies.

Beginning in 1985, a program to map the northernpart of the Ouachita Mountains fold-and-thrust belt and thesouthern part of the Arkoma foreland basin was conducted.Designed to support natural-gas exploration and coaldevelopment, and to reduce associated environmental haz-ards, 22 geological-quadrangle maps were produced. In1998, OGS shifted new mapping efforts to the northernOklahoma City metropolitan area. Twelve 7.5-minutequadrangles were selected, with completion expected byJuly of the year 2000. The purpose of this mapping is toprovide area planners with detailed geologic maps that willenable them to make informed decisions with regard toaquifer protection, resource development, and highwayconstruction.

Geologic Mapping at the Oklahoma Geological Survey:The Move From Traditional to Digital Cartography

By T. Wayne Furr

Oklahoma Geological SurveyNorman, Oklahoma 73019-0628

Telephone: (405) 325-3031Fax: (405) 325-7069



In late 1994, the Oklahoma Geologic MappingAdvisory Committee (OGMAC) recommended that OGSprepare a series of geologic maps at a scale of 1:100,000for the entire State using digital technology. The purposeof the maps is to provide a Geographical InformationSystem (GIS) geologic data base for industry, public offi-cials, area planners, and other interested parties. In addi-tion, this series will provide the foundation for a new1:500,000-scale geologic map of the State. Maps in thisseries are compilations of geologic investigations fromvarious sources, with field checks used to fill gaps foundin prior investigations and improve on earlier mappingefforts.

Partial funding for the OGS geological mapping activ-ities has been provided under two separate grants. From1985 to 1993, mapping was funded through the COGE-OMAP Program, an agreement between the OGS, theArkansas Geological Commission, and the USGS. Since1994, geologic mapping has been funded through theSTATEMAP component of the National CooperativeGeologic Mapping Program.

1:100,000-SCALE DIGITAL COMPILATION

In 1996–1997, two 30 x 60 minute quadrangles inwestern Oklahoma were compiled to test OGS compilationefforts and digital capabilities. The Watonga and FossReservoir quadrangles (figure 1) were chosen because thebedrock geology, consisting mostly of gently dippingPermian redbeds, is relatively simple and well known.The highly dissected terrain, however, results in complexcartographic patterns.

In September 1996, OGMAC recommended that OGSconcentrate on the Oklahoma Panhandle. The mappingwill complement ongoing studies in the area conducted bythe Oklahoma Water Resources Board and WaterResources Division of the USGS. Investigations by eachagency will provide a better understanding of environmen-tal issues associated with an increasing transportationinfrastructure, numbers of feed lots, and meat-processingplants. OGS geologists started compiling the geology inthe westernmost quadrangle in 1997. The digitizing, GIS-attributing, and cartographic production has essentiallybeen completed on the Boise City Quadrangle.Completion of the Guymon and Beaver quadrangles isexpected by the end of June 1999 (figure 1).

Methods Used

I. OGS Geologist1. Conducts library research and compiles all existing

modern geologic contacts on a 1:100,000 green-linebase (Note: A green-line base is a frosted-film sheetwith the base map printed in green).

2. Supplements compilation and resolves different geolog-ic maps with air-photo interpretation and/or reconnais-sance field checking.

3. After compilation, a paper print is made and the geolo-gist colors the formations, checking for gaps or openformation contacts.

4. An explanation, area of investigation map, and othernecessary information is prepared.

5. This package of information is turned over to the car-tographer for processing.

II. OGS Cartographic Staff

6. Using scribing, the cartographer makes a line-workseparation of the geologic contacts from the base map.

7. Prepares a geologic color-selection guide

8. Prepares a map layout guide.

9. Makes a clear film or paper photo print of the geologicline work.

10. The photo print, color selection guide, and map layoutguide is turned over to the GIS specialist. Currently,the GIS processing is under contract to an individualat Oklahoma State University (OSU).

III. GIS Specialist

11. The specialist scans the 1:100,000-scale photo positiveof the geologic map sheet at 400 dots per inch on anANAtec 3640 Eagle optical scanner.

12. The scanned (raster) image is converted to vectorpolygons utilizing LtPlus raster-to-vector conversionsoftware on a SUN SPARC workstation.

13. A plot of the digital vector data is made at 1:100,000scale on an HP650C plotter and visually comparedwith the original map compilation to ensure complete-ness and precision of scanning and data conversion.

14. Within LtPlus, polygon topology is created, each poly-gon is attributed, and a series of quality-controlchecks is performed with software macros (programs).

15. The digital data is then exported from LtPlus in stan-dard USGS DLG-3 format and imported into Arc/Infoversion 7.0.3.

16. A plotting routine is written and executed in ArcMacro Language program to utilize the digital geolo-gy polygons as part of a 1:100,000-scale geologicmap.

17. Base map layers, a title, explanation, text, and indexmaps are added to complete the map layout.

18. On request, the completed map data sets will be madeavailable to the public in one of the following formats:(a) Arc/Info export format (.e00), (b) ArcView shape-files, or (c) USGS DLG-3 format for access through

the World Wide Web via a Web browser or by filetransfer protocol (ftp) (Furr, Gregory, and Suneson,1998). Note: At the present time, the exact proce-dure, file format, and method of distribution describedin step 18 have not been determined.

GIS vs. Published Maps

In addition to the GIS format, Dr. Charles J. Mankin,Director of OGS, expressed the need to publish each geo-logic quadrangle in full color for release in our GeologicMap Series. Having two different types of digital productspresented a different set of problems. The biggest com-plaint expressed by reviewers of printed GIS maps wasgraphic presentation. They expected to see standard geo-logic colors, lettering, and other symbols. Unfortunately,graphic presentation has never been a strong point in mostGIS programs. Another problem was converting the GISfiles to a publishable format. A GIS is an analysis tooldesigned to evaluate a range of possible scenarios. Byevaluating different possibilities, a course of action can beconsidered before irrevocable mistakes are made in thelandscape itself (Burrough, 1986, p. 7). In other words, aGIS was never intended to be the computer toolbox fordesigning, drafting, and printing maps in the traditionalway. As a result, from the GIS file format, it is difficult to

provide to the printing industry a digital file that is com-patible with the industry’s graphic-output devices.

These problems and others have been addressed byagencies that have tried to use a GIS as a publishing tool.In checking with other state geological surveys, privatecartographic firms, pre-press, and printing companies, theanswers are not entirely clear. Many agencies useMacintosh computers, others use PCs with a variety ofoperating systems, and others use UNIX-based worksta-tions. A multitude of software was also found to be in use.Some were using one of several GIS programs, some wereusing CAD programs, while others were using a variety ofgraphic software programs. The pre-press and printingfirms were not familiar with GIS file formats, but pre-ferred to use files created in Adobe Illustrator.

The OGS Cartographic Section’s computer setup con-sists of Pentium II PCs with Windows NT operating sys-tems. Software includes Microsoft Office, ArcView 3.0a,and Adobe Illustrator 7.0.1. One problem that had to beaddressed was how to import GIS files to Illustrator whilemaintaining the GIS attributing. For this operation, third-party software was needed to bridge the gap betweenArcView shapefiles and Adobe Illustrator. The third-partysoftware chosen was MAPublisher from Avenza Software.MAPublisher is a cartographic-geographic informationsystem for integrating GIS files directly into AdobeIllustrator while maintaining the GIS attributing. The

GEOLOGIC MAPPING AT THE OKLAHOMA GEOLOGICAL SURVEY 9

0 50 Miles

0 50 Kilometers

FossReservoir

Watonga

Boise City Guymon Beaver

CONVENTIONAL MAPS

1:24,000 (completed)

1:24,000 (in progress)

1:24,000 (planned 2000)

Coal Resource Maps (completed)

County Map (completed)

County Map (in progress)

1:100,000 DIGITAL MAPS

Completed

In Progress

Planned 2000

OTHER DIGITAL MAP PROJECTS(in progress)

Oil & Gas Map of Oklahoma

Geologic Provinces of Oklahoma

Buffalo

Woodward

Figure 1. Index to 1:100,000-scale geologic maps in Oklahoma.

plug-in filters allow the cartographer to consider map pro-jections, scale, color, and necessary cartographic opera-tions, while maintaining the GIS functionality. The com-pleted map project can be saved in one of the many for-mats used by the pre-press and printing industry. In addi-tion, map layers can be exported as shapefiles for use in aGIS.

Adjustments to OGS Methods

As a result of the lessons learned from the Watongaand Foss Reservoir quadrangles and the availability of newsoftware, several changes in the method of map productionwere made as work progressed on the Boise City quadran-gle. Steps 1 through 16 described previously remained thesame. One change being considered for future projects isdropping the printed green-line base in favor of a clear-film base map with a frosted-mylar overlay. The expectedgoal is that the geologist carefully drafting the geologiccontacts will eliminate the need for the cartographer tomake a line-work separation and the expense of a photo-graphic positive.

Adjusted Methods

To recap step 16 from before: A plotting routine iswritten and executed in Arc Macro Language program toutilize the digital geology polygons as part of a 1:100,000-scale geologic map.

17. The geologic-polygon and base map image file layersare placed on the OSU (Department of Plant and SoilSciences) file server in ArcView shapefile format.

18. The cartographer transfers the files to the OGS com-puter network using standard file transfer protocol(ftp)

19. The ArcView shapefiles are moved into AdobeIllustrator using MAPublisher’s import filter.

20. Using Illustrator, the cartographer performs the neces-sary cartographic work to bring the map to OGS stan-dards by adding the title, explanation, index map(s),geologic letter symbols, and colors.

21. File formats for printing the maps are created and sentto the printer.

22. The geologic map data base is exported in ArcViewshapefile format for use in GIS applications.

23. On request, the completed map data sets will be madeavailable to the public in one of the following formats:(a) Arc/Info export format (.e00), (b) ArcView shape-files, or (c) USGS DLG-3 format for access throughthe World Wide Web via a Web browser or by filetransfer protocol (ftp). Note: At the present time, the

exact procedure, file format, and method of distribu-tion described in step 23 have not been determined.

SUMMARY

The OGS Cartographic Section started using comput-ers to produce figures and illustrations for publicationsabout five years ago. Geologic maps in the 1:100,000-scale series are the Survey’s first attempt at using GIS con-cepts and graphics software jointly to produce maps forprinting. The methods used provide a workable solutionthat can be adopted by agencies with similar productionneeds. Cartographic production of the Boise City map wasapproximately three times faster than traditional draftingmethods. The need for cartographic materials, such asscribe-sheets, peel-coats, type-overlays, and other tradi-tional materials, as well as the use of supporting laborato-ries is virtually eliminated. The various geological map-ping programs at OGS are expected to expand as experi-ence is gained using GIS in combination with digital car-tography. To accomplish overall goals, methods will beadjusted and production speed is expected to increase.With materials and support for traditional cartographicmethods diminishing the need to totally convert to digitalproduction for all maps at the OGS is rapidly approaching.

Future mapping will include additional 7.5-minutequadrangles in the Oklahoma City area and 1:100,000-scale quadrangle(s) in western Oklahoma. Additional pro-jects being considered for production in both digital andpublishable formats are the Oil and Gas Map of Oklahomaat a scale of 1:500,000, the Geologic Province Map ofOklahoma at a scale of 1:750,000, and the expanded revi-sion of Educational Publication 1. In addition to the geo-logic mapping programs, an inventory of all geologic mapspublished by OGS is being compiled and entered into theNational Geologic Map Database<http://ngmdb.usgs.gov>. As a state agency, theOklahoma Geological Survey has made a commitment tomeet the needs of the State’s citizens by providing high-quality printed maps, as well as GIS data bases of theState’s geological resources.

REFERENCESBurrough, P.A., 1986 (10th printing with corrections, 1993),

Principles of Geographical Information Systems for LandResources Assessment: Oxford University Press, Inc., NewYork, 194 p.

Furr, T.W., Gregory, M.S., and Suneson, N.H., 1998, OklahomaGeological Survey’s (OGS) digital geologic maps (abs.):Geological Society of America Abstracts with Programs, v.30, no. 3, p. 6.

Ham, E.A., 1983, A History of the Oklahoma Geological Survey1908–1983: Oklahoma Geological Survey SpecialPublication 83–2, 60 p.


11

INTRODUCTION

During the transition from manual to digital carto-graphic preparation of thematic maps over the past severalyears the Central Publications Group (CPG) of theGeologic Division, U.S. Geological Survey (USGS), hasproduced maps from a variety of compilation materialssubmitted by authors: conventional drafting on film; mapsdigitized by authors or contractors in GSMAP or, morerecently, GSMCAD (both programs available at<http://ncgmp.cr.usgs.gov/ncgmp/gsmcad/> or<http://greenwood.cr.usgs.gov/maps/software.html>; for adescription, see Williams, 1997); maps digitized byauthors or contractors in Arc/Info; maps drafted by authorsin a graphics program, such as Adobe Illustrator; mapsgenerated by programs widely used by USGS geophysi-cists; and hybrid maps using more than one of theseapproaches. Some authors produce final cartographic lay-outs for their maps themselves, and others rely on CPG forfinal production. Along the way we have encounteredproblems with digital production from these varioussources but have found solutions to the extent that most ofthe maps currently in production are wholly digital in thesense that the map is output from a digital file.

In addition to the challenge of producing maps fromsuch a variety of compilation materials, we need to pro-duce maps for release online, for plotting on demand on ahigh-resolution plotter, and for printing on a press; we alsoneed to work with authors to place digital databases andmetadata online. Our poster presentation for this work-shop outlines some procedures for producing maps thatwork for us in our publications environment.

PRODUCTION PROCEDURES

We import digital files for geologic maps into AdobeIllustrator via Avenza MAPublisher and use the powerfulgraphics capabilities of Illustrator to compose the final lay-out of a map sheet rather than trying to do so in a GIS (seefig. 1). Composing the map sheet in this way lets us speedup production dramatically while retaining the traditionallook of a USGS map.

Step 1. Digitize the Map

We obtain digital files from the author or we pay acontractor to do the work. The geologic maps referencedhere and in our poster display were digitized in a CADprogram or scanned and processed into Arc/Info cover-ages.

Before contracting for the digitizing we sometimeschoose first to have a cartographic contractor scribe verycomplex conventionally drafted linework. The contractorscribes one sheet of color boundaries and one or two(depending on complexity) other sheets of lines (such asfaults, folds, or caldera boundaries), using a uniform lineweight of 0.006 inches and omitting line decorations andpoint symbols. In some cases we have the contractor inkthe map unit symbols on an overlay. We then have clear-film positives made from the scribes and furnish thesematerials to the digitizing contractor for scanning.

If the linework is not too complex, inked compilationson film are scanned directly. Whether separates arescribed or the original compilation is scanned, we ask theauthor to make color-coded line and symbol guides to help

Digital Geologic Map Production and Database Development inthe Central Publications Group of the Geologic Division,

U.S. Geological Survey

By Diane E. Lane, Alex Donatich, F. Craig Brunstein, and Nancy A. Shock

U.S. Geological SurveyBox 25046

Denver Federal Center, MS 902Denver, CO 80225

Telephone: (303) 236–5476Fax: (303) 236–6287


the contractor accurately attribute polygons, lines, andpoint symbols. Point symbols on overlays or on the origi-nal compilation on film are either (1) digitized by hand or

(2) scanned and then used on-screen as a guide for digitiz-ing point symbols in a graphics program. In some caseswe provide the digitizing contractor with scans of the


Digitize the map GIS files

Import files into Illustratorvia MAPublisher

Write metadataARC/INFO shapefiles

DXF files

GIS files andmetadata placed onpublic-access server

Stroke lines andfill map units with colors

and patterns

Draft or importsupplementalillustrations

Scan a clear-filmpositive of the

topographic base

TIFF image

Import the TIFF into Photoshop, change from grayscale to bitmap,

and crop image

Register TIFF image in ARC/INFO in required projection

Place TIFF image in Illustratorlayout and register to other layers

via MAPublisher

Import text from a word processor into linked type containers

and add miscellaneous type

Color-separated negativesfor printing on a press

OR

RIP file and metadatafor map-on-demand

PDF file for releaseonline

Author compiles map

Figure 1. Chart showing basic procedure for digital map production and database development in the CentralPublications Group of the Geologic Division, U.S. Geological Survey

topographic base, especially if needed to digitize open-water boundaries.

In contracting for digitizing, we provide standards foraccuracy and for coverage attributing. In order to meetour need for both a digital database and usable graphicsfiles, we request two files to represent point symbols. Oneis an attributed Arc/Info point coverage, and the other is aline coverage that graphically represents the point sym-bols. Symbols such as the bar and ball on faults or dis-placement arrows on folds are represented as points (onthe line) and attributed accordingly. In other words, we donot want such symbols to be evenly spaced along a line asa decoration such as sawteeth would be, but to be placedexactly where the author placed them.

Step 2. Import Files into Graphics Programvia MAPublisher

Avenza MAPublisher 3.0 is a suite of plug-in filtersfor Adobe Illustrator and Macromedia Freehand that allowthese formats to be imported with all attributes intact:Arc/Info generate, ArcView shapefiles, DXF, MapInfomid/mif, DLG, and SDTS. For ArcView shapefiles, keep-ing the map attribution intact means that like-attributedfeatures can be selected as a group via MAPublisher inorder to adjust the graphic portrayal of that feature. Forexample, lines that were attributed as solid contacts can beselected as a group in the graphics program, and lines thatwere attributed as dashed or dotted contacts can be select-ed as a group and assigned the proper line weight andlength of dash. Similarly, the polygons of a map unit canbe selected as a group in order to assign colors and pat-terns. For DXF files exported from a CAD program suchas GSMCAD, like-attributed features are imported intotheir own layers in the graphics program.

For maps digitized in a CAD program, DXF filesimport properly for polygons, lines, point symbols, andtext. For maps from Arc/Info, we import shapefiles for thepolygons and lines but use DXF files for the text and forgraphical representations of point symbols. In CPG wehave worked most extensively with ArcView shapefilesand DXF files. In importing a set of files for a map, it isimportant not to move the objects imported into Illustratorvia MAPublisher until all files have been imported inproper registration.

Step 3. Compose Map Sheet Layout inGraphics Program

Once the digital files for the map have been imported,the line elements of the geologic map itself are stroked(assigned a line type and weight) and polygons, text, andsome point symbols are filled with colors and (or) patterns.To select all polygons for one map unit or all instances ofa concealed contact, for example, we select all art on onelayer in Illustrator or we use the “select by attribute” func-

tion of the MAPublisher filter. Which approach we usedepends on whether the imported elements were DXF filesor ArcView shapefiles.

Cross sections and other supplemental illustrationsmay be digitized in a CAD program or in Arc/Info, or theymay be drafted in a graphics program. GIS files for theseillustrations may be imported into the graphics program inthe same manner as the map itself. We generally importthese into their own Illustrator document to keep the work-ing file size small and adjust lines and colors as we do onthe map itself. Layers for these supplemental illustrationscan be preserved by checking “paste remembers layers” inIllustrator when copy-pasting the illustration into the lay-out. However, you will not be able to use the “select byattribute” function of MAPublisher to edit an illustrationthat is no longer in the Illustrator document into which itwas imported.

In choosing colors for a map, we consider whether themap will be printed on a press or only on a plotter. In bothcases we follow USGS standards for assigning colors byage and rock type. We choose colors from standard sheetsof printed process colors for maps that will be printed on aprinting press, but we use a plotted version of the colorchart to refine color choices for maps that will be printedonly on a plotter. Since plotters differ in their output, weuse color charts printed on the same kind of plotter as thatused to print maps in the maps-on-demand series for saleto the public.

Using Illustrator, everything that can be done in thenondigital production of a map can be duplicated, includ-ing adding patterns to color-filled polygons. We haveavailable in digital form many of the patterns traditionallyused on USGS thematic maps. To apply the patterns, wecreate a separate layer in Illustrator for each map unit to bepatterned. We select the polygons for each of these mapunits and copy them to the new layer, then fill them withthe pattern. The polygons on this layer should have nostroke, only the pattern fill. The polygons that are filledwith the background color also have no stroke, only thecolor fill, and should be the bottommost layer in the layerspalette in Illustrator. The next lowest layer would betopography, if present, and the layers of polygons filledwith patterns should be above the topography. We havefound that this order is suited to both printing on a pressand on a plotter.

The rest of the map sheet layout is generally com-posed in the Illustrator document into which the map wasoriginally imported. A typical USGS map includes a cor-relation of map units, a description of map units, discus-sion, references, and supplementary illustrations or pho-tographs. Any cross sections or other illustrations draftedin Illustrator or available in other formats (such as EPS orTIFF) are copy-pasted or placed in the Illustrator layout,and text is imported from a word processor (such as MSWord) into linked type containers. Special characters(such as em dashes and degree symbols) are usually pre-

DIGITAL MAP PRODUCTION AND DATABASE DEVELOPMENT IN THE CENTRAL PUBLICATIONS GROUP 13

served, but formatting such as hanging indents and fontselections is not—this formatting is done in Illustrator.The addition of bar scales, titles, headnotes, figure cap-tions, and so on completes the layout except for the topo-graphic base.

Step 4. Add a Topographic Base

If a satisfactory vector base (digital line graph) is notavailable, a scale-stable positive of the base can bescanned and saved as a grayscale (8-bit) TIFF image. Wecan open the TIFF image in Adobe Photoshop, adjustthreshold levels to preserve detail, change the mode fromgrayscale (8 bit) to bit map (1 bit), and crop the image asneeded. We can then register the TIFF image in Arc/Infoin a projection that matches the digital coverages for themap, place the TIFF image on its own layer in theIllustrator layout, and use MAPublisher’s “register image”filter to reference and scale the image. This new layer isplaced directly above all of the color-filled polygons butbelow any layers of pattern-filled polygons.

In printing geologic maps on a press, we have custom-arily screened topographic base maps to 30–40 percentblack. However, output from a plotter may be more satis-factory if the topographic base layer is screened to 50–60percent black and if color fills of map units (except forunits that occupy only small areas) are no more than 40percent cyan or magenta (higher percentages of yellowmay be used). A base layer screened to a percentage ofblack that is less than the percentages of the underlyingcyan or magenta in larger map areas appears to block outthe underlying color rather than to overprint it (for exam-ple a 30-percent black topographic base will appear toblock out the color of a map unit that is 40 percent magen-ta).

Step 5. Output the Map Sheet

We have several ways of outputting a digital map. (1)Print on paper at a printing plant. This approach entailsthe expense of making color-separated negatives from thedigital file and then printing and storing the maps. (2) Plota copy as needed. Plotting on demand eliminates the needto make color-separated negatives and to store the printedmaps. However, each on-demand RIP (acronym for RasterImage Processor) file must be archived on a CD-ROM.(3) Convert the digital file to a PDF file and release itonline (maximum page size in the newly released AdobeAcrobat 4.0 is 200 x 200 inches).

For maps printed on a press, we currently use a ser-vice bureau to output color-separated negatives from ourdigital files or we give the digital files directly to the print-ing contractor. In the latter case the printing contractor isresponsible for obtaining color-separated negatives fromthe files we provide. Two examples of maps output in this

way for printing are Geologic Investigations Series I–2627(Day and others, 1998) and I–2630 (O’Neill, 1998).

In CPG we are producing more and more maps thatwill be “printed” only on a plotter. We export theIllustrator file as an EPS and use Onyx PosterShop ClientServer, v. 4.5, to make a RIP file. We provide the USGSBranch of Information Services with the input file (usuallythe Illustrator file), the RIP file, a README file, andmetadata together on a CD-ROM. When a customerorders a map-on-demand product, the RIP file is then plot-ted on heavy coated paper on an HP Design Jet 3000 CPplotter using UV-resistant inks. Three examples of mapsoutput in this way for the map-on-demand system areGeologic Investigations Series I–2652 (Winkler and oth-ers, 1999), I–2656 (Lidke, 1998), and I–2667 (O’Sullivan,1998). Metadata and information about ordering thesemaps are available at <http://rmmcweb.cr.usgs.gov/public/mod/>.

Whether we print the map on a press or on a plotter,we now convert the Illustrator file to a PDF file and placeit on a public-access server as a complement to the hard-copy map. Some examples of maps output as PDF filesare Geologic Investigations Series I–2627 (Day and others,1998), I–2630 (O’Neill, 1998), I–2631 (O’Neill and Nutt,1998), I–2652 (Winkler and others, 1999), I–2656 (Lidke,1998), and I–2667 (O’Sullivan, 1998). The PDF files forall these maps may be accessed through<http://greenwood.cr.usgs.gov/maps/maps.html>.

DEVELOPING AND SERVING A GISDIGITAL DATABASE FOR THE MAP

We have started placing GIS digital files and metadatain support of the printed or plotted map on a public-accessserver. A README file describes the available files, andmetadata written to the Content Standards for DigitalGeospatial Metadata published by the Federal GeographicData Committee (FGDC) give keywords and detailedinformation about the scope of the mapping project, thecoverages (including descriptions of entities and attribut-es), the production process, and contacts for further infor-mation. One example (Skipp and others, in press) may beaccessed at <http://greenwood.cr.usgs.gov/pub/I-maps/I-2634/>.

We carefully review the files provided by the contrac-tor to ensure that all digitizing and attributing errors arecorrected in the GIS before placing the GIS files onlineand importing the files into the Illustrator layout. Even ifa printed or plotted map is available, the database may dis-play some elements differently than shown on the hardcopy. Therefore, we explain the features of the database asthoroughly as possible in the metadata, pointing out anydifferences that might be critical to the user.


CONCLUSIONS

The advantages we see in this method of map produc-tion are timeliness, ease of distribution, and adaptability ofthe digital files for different modes of output. Dependingon the complexity, composing a map sheet in this waytakes only a few days to a week or so once all the elementsof the map sheet are in digital form. The completely digi-tal format makes minor revision and reissue of the mapquick and easy. Maps produced in this way can be outputreadily in the cost-effective maps-on-demand system andcan be released online.

ACKNOWLEDGMENTSThis paper is an expansion of a report prepared by

Alex Donatich to describe map production procedures thathe initiated in the Central Publications Group. NancyShock has been invaluable in integrating GIS with graph-ics procedures.

REFERENCES CITED

Day, W.C., Dickerson, R.P., Potter, C.J., Sweetkind, D.S., SanJuan, C.A., Drake, R.M., II, and Fridrich, C.J., 1998,Bedrock geologic map of the Yucca Mountain area, NyeCounty, Nevada: U.S Geological Survey GeologicInvestigations Series I–2627, scale 1:24,000; PDF fileaccessible through URL<http://greenwood.cr.usgs.gov/maps/covers/i-2627/i2627.html> [map displays as one PDF file of 10.1MB].

Lidke, D.J., 1998, Geologic map of the Wolcott quadrangle,Eagle County, Colorado: U.S. Geological Survey GeologicInvestigations Series I–2656, scale 1:24,000; PDF fileaccessible through URL<http://greenwood.cr.usgs.gov/maps/maps.html#I-MAPS>[map displays as two PDF files of 2.4 MB each].

O’Neill, J.M., 1998, Geologic map of the Cienega School quad-rangle, Otero County, New Mexico, and Hudspeth County,

Texas: U.S. Geological Survey Geologic InvestigationsSeries I–2630, scale 1:24,000; PDF file accessible throughURL <http://greenwood.cr.usgs.gov/maps/covers/i-2630/i2630.html> [map displays as one PDF file of 1.9MB].

O’Neill, J.M., and Nutt, C.J., 1998, Geologic map of theCornudas Mountains, Otero County, New Mexico: U.S.Geological Survey Geologic Investigations Series I–2631,scale 1:24,000; PDF file accessible through URL<http://greenwood.cr.usgs.gov/pub/i-maps/i-2631/i2631.html> [map displays as one PDF file of 2.3MB].

O’Sullivan, R.B., 1998, Correlation of Jurassic San Rafael Groupand related rocks from Blanding, Utah, to Dove Creek,Colorado: U.S. Geological Survey Geologic InvestigationsSeries I–2667; PDF file accessible through URL<http://greenwood.cr.usgs.gov/pub/i-maps/i-2667/index.html> [maps displays as one PDF file of 155kb].

Skipp, Betty, Lageson, D.R., and McMannis, W.J., in press,Geologic map of the Sedan quadrangle, Gallatin and ParkCounties, Montana: U.S. Geological Survey GeologicInvestigations Series I–2634, scale 1:48,000; databaseaccessible online at URL<http://greenwood.cr.usgs.gov/pub/I-maps/I-2634/>.

Williams, Van, 1997, Using the GSMCAD program with GPS fordata collection in the field and as a quick and efficient wayof creating Arc/Info geologic map coverages, in Soller,D.R., ed., Proceedings of a workshop on digital mappingtechniques—Methods for geologic map data capture, man-agement, and publication: U.S. Geological Survey Open-File Report 97–269, p. 83–87; online version accessible atURL <http://ncgmp.usgs.gov/pubs/of97-269/williams.html>.

Winkler, G.R., Goldfarb, R.J., Pickthorn, W.J., and Campbell,D.L., 1999, Maps showing areas of potential for metallicmineral resources in the Valdez 1x3-degree quadrangle,Alaska: U.S. Geological Survey Geologic InvestigationsSeries I–2652, scale 1:250,000; PDF file accessible throughURL <http://greenwood.cr.usgs.gov/maps/maps.html#I-MAPS>[map displays as one PDF file of 7.4 MB].

DIGITAL MAP PRODUCTION AND DATABASE DEVELOPMENT IN THE CENTRAL PUBLICATIONS GROUP 15

17

ABSTRACT

Avenza Software Inc., the developers of MAPublisher,is a small company located in Burlington, Ontario.MAPublisher is a suite of Plug-Ins and Xtras for AdobeIllustrator and Macromedia FreeHand. MAPublisher addsGIS capabilities to these two graphics applications. It addsgraphics capabilities that are unavailable with a typicalGIS. MAPublisher is used by government agencies, edu-cational institutions, and businesses worldwide to createpublication quality mapping products from GIS datasources.

AVENZA

The founder of Avenza was hired by the graphicsdepartment of a major company to make maps. He quick-ly realized that he was spending hours completing workthat was already available. Why scan and screen digitizeendless sheets of paper maps when accurate and georefer-enced vectors were readily available? Why spend hoursadding database attributes to vectors when this data hadalready been collected? In 1995, along with a GIS pro-grammer and two other partners, he realized the need for abetter way of making maps. This realization led to thecreation of Avenza and its main product, MAPublisher.Avenza is now a small but innovative company employingabout 10 people. The staff includes an engineer, a webdesigner/system administrator, a cartographer and a coregroup of programmers. Avenza is located in Burlington,Ontario, on the outskirts of Toronto.

What is MAPublisher?

MAPublisher is a suite of Plug-Ins and Xtras forAdobe Illustrator and Macromedia FreeHand that bridges

Geographic Information System (GIS) technology withhigh-end graphics software for high resolution printing andelectronic publishing technology. Cartographic qualitymap production is now faster, easier, and better. Avenzaunderstands that GIS graphic tasks are best performed inthe right environment such as powerful illustration pro-grams like Macromedia Freehand and Adobe Illustrator.MAPublisher takes you into this environment seamlesslyand effortlessly with the right GIS data management toolsto facilitate the map production process. Using this fast,intuitive system, your map can transcend the ordinary andbecome a work of art.

Why Use MAPublisher?Graphics Functionality Unavailable WithYour GIS

MAPublisher gives you the ability to perform GIStasks in a graphics environment, and allows you to addgraphic attributes unavailable with most GIS products onthe market. With MAPublisher and Adobe Illustrator orMacromdia FreeHand, you have

* Powerful graphics tools for total control of cartographicvisualization.

* Unlimited colour choices including 24 bit colour and theability to print CMYK four-colour separations.

* The ability to attach text to paths to create natural flow-ing labels.

* Complete typeface, point size, leading, kerning, trackingand alignment control and the ability to embed text.

* Minimal pre-press and printing problems.

* Automated crop and trim mark creation.

* Trapping for compensation of misregistration.

Avenza’s MAPublisher

By Susan Muleme

Avenza Software, Inc.3385 Harvester Road, Suite 205

Burlington, Ontario, Canada L7N 3N2Telephone: (905) 639-3330



* The ability to convert custom colours to process coloursfor printing.

* And all of the other features of your graphics program!

What can I do with MAPublisher?

* Import the most widely used GIS data files (Generate,Shapefile, MID/MIF, DXF, DLG, SDTS).

* Export data Shapefiles, MID/MIF or PDF’s.

* Create “Smart” PDF’s with queriable databases withpdfPLUS (another Avenza product).

* Import, merge, create and edit any database files.

* View a floating window of map attribute information.

* Assign, edit, analyze and query map attributes.

* Tile map files together in real world coordinates.

* Automate raster image registration of DRG’s and othertypes of georeferenced raster image.

* Automate your legend creation and labeling.

* Change map projections with the fully customizable pro-jection editor (has 119 projections and over 40 ellip-soids).

* Quickly identify geographic locations and display themap scale and anchors in a floating window.

Who Uses MAPublisher?

MAPublisher is used by professionals in a wide vari-ety of fields including geology, engineering, forestry,municipal planning, mining and exploration, graphicdesign, advertising, real estate, law, defense, newspapers,and television media. Many municipal, provincial, state,federal, and international governments use this product toproduce maps and graphics; evidence is provided in talksdelivered at this meeting. MAPublisher is being used toteach cartographic concepts at educational institutionsaround the world. For example, the University ofWaterloo uses MAPublisher to teach projections to its stu-dents; ITC in the Netherlands uses MAPublisher in combi-nation with ILWIS to teach geomatics concepts from GISto remote sensing imagery analysis.

CONTACTING AVENZA:

Please visit Avenza’s website,<http://www.avenza.com>, for more information aboutMAPublisher. For examples of MAPublisher maps, checkin the Products-Demos section of the web page. If youhave any technical or sales questions about MAPublisher,or if you would like to book a training session, you cancontact Avenza in a number of ways:* Sales Department and Order Desk [[email protected],

or tel. (800)884-2555, ex.50]* General Information [[email protected] or tel. (905)

639-3330]* Technical Support [[email protected] or tel. (905)

639 2329]

19

HP’s Thermal Inkjet Technology (HP TIJ) has beenadapted for use in large format printing by the BarcelonaDivision. TIJ technology has provided high levels ofmonochrome and color print quality, speed, and reliabilityat very low cost. TIJ is a relatively new printing technolo-gy with lots of life left in it. Performance levels are con-stantly being increased to meet the emerging needs oftechnical and graphics printing applications. Users ofGeographical Information Systems (GIS) have particularlydemanding requirements.

In my talk, I will mention some of the trends shapingGIS, especially the printing part, and how we at HP’s

Barcelona Division are using those trends to guide us indeveloping HP TIJ technology. I will highlight the chal-lenges in developing TIJ writing system architectures, talkabout future directions, and explain how we collaboratewith applications developers such as ESRI and AdobeSystems in our effort to provide more complete solutions.

For information on HP DesignJet products, pleaseconsult:

<http://www.hp.com/go/designjet>.For HP DesignJet post-sales information and other

services, please consult:<http://www.designjet-online.hp.com>.

DesignJet Large Format Printer Trends and Technologies

By Alberto Berry

Hewlett-Packard CompanyAvenida Graells, 501

08190 Sant Cugat del VallesBarcelona, Spain

Telephone: +34 93 582 27 38Fax: +34 93 582 29 35


21

ABSTRACT

Digital base map data of modern geologic maps mustconform to the requirements of the National Spatial DataInfrastructure by containing the following framework datalayers: transportation, hypsography (elevation data),hydrography, political units (boundaries), geodetic control,and a cadastral reference system (e.g., PLSS). At the onsetof digital mapping a decade or so ago, overall quality ofdigital base data was quite poor compared to the non-digi-tal, photographic plates previously used in traditional car-tographic methods, and it has never adequately improved.These digital geographic base data are currently availablefrom the USGS in four basic formats: DOQs, DEMs,DLGs, and DRGs. DOQs and DEMs are best utilized inconjunction with DLGs or DRGs due to a lack of severaltypes of framework data. Only the DRG contains all nec-essary framework elements, yet it is not available in highresolution. A case study from a high relief area with asmall contour interval clearly illustrates the overall poorquality of this scanned data. All four datasets requireextensive modification prior to incorporation into the geo-logic map. The best solution at present is to generate in-house DRG data by scanning photographic plates at muchhigher resolution and rubber-sheeting the raster imagearound the sixteen basic grid control points in Arc/Info.

INTRODUCTION

A geologic map is of little use to anyone if it is not insome manner “grounded” to its appropriate location on theearth by registering it to the topographic base map. At thevery least, it must be projected to a known geospatial coor-dinate system (Universal Transverse Mercator, or UTM,

latitude and longitude, etc.), perhaps with an adjacentnorth arrow, before any scientific editor would consider itacceptable for publication. More importantly, I wouldargue that the geologic map of the late 1990s should be incompliance with the National Spatial Data Infrastructure(NSDI) and contain essential thematic framework basedata on which information (geologic or otherwise) can beaccurately assimilated, registered, and integrated. Of theeight themes of geographic data recognized by the FederalGeographic Data Committee (FGDC) as “theFramework,” the base of any geologic map should con-tain the following layers: transportation, hypsography (ele-vation data), hydrography, political units (boundaries),geodetic control, and a cadastral reference system (e.g.,PLSS). Digital orthoimagery often provides a dramaticbackdrop upon which to display geologic data if the dataaren’t too “busy.” Cadastral land ownership, of all of theframework elements, seems the least important to the geo-logic mapper. However, in New Mexico, where Indianpueblos and sovereign tribal lands make up almost 20% ofthe total land area of the fifth largest state of the Union,this is significant geographic data that should be included.

While all of this might seem implicit and ratherunnecessary to raise as an issue in a forum of geologicalmapping professionals, I feel that, to date, this subject mat-ter has not been adequately addressed. In the days of pre-digital cartography, base materials were most oftenacquired as photographic negative plates from the USGSand then manually manipulated by the cartographer to fitthe map as needed. Little thought was given to the typesof data presented. The requirement of the base was simplyto provide an adequate means of location. With the digitalrevolution, cartographers quickly began utilizing a widevariety of hardware and software to expedite the map-making process. While in-house map production time was

“Can’t See the Geology for the Ground Clutter” —Shortcomings of the Modern Digital Topographic Base

By David J. McCraw

New Mexico Bureau of Mines and Mineral ResourcesNew Mexico Institute of Mining and Technology

801 Leroy PlaceSocorro, New Mexico 87801Telephone: (505) 835-5487



greatly reduced, problems involving file and data sharinggrew considerably. Indeed, in response to this, the NSDIwas established in 1994 by President Clinton’s ExecutiveOrder 12906.

At the onset of digital mapping, overall file size wasperhaps the most important consideration in practical digi-tal map production. Base data quality suffered greatly inan attempt to keep file sizes manageable, when comparedto those non-digital base data previously utilized. Thiswas an acceptable trade-off to the overwhelming amountof time and effort saved. Today, with advanced technolo-gy and faster computers, base data quality continues tosuffer, unnecessarily. Base data underpinning geologicmaps commonly lack one or more framework elements,are often illegible (either due to poor resolution or fromhaving type or linework set over base information), inac-curate (due to outdated data, a lack of geodetic control, amisregistration of the base data, or some combinationthereof), or both. Problems of this nature often result fromthe inadequate quality of the digital data available. At pre-sent, all digital base data require some level of modifica-tion to comply with NSDI prior to adding geology. In thispaper, I compare and contrast these various digital dataand highlight some of their shortcomings.

TYPES OF DIGITAL TOPOGRAPHICBASE DATA AVAILABLE

Unless geologic cartographers intend to produce theirown base data through extensive GPS-surveying and/ormanipulating some other remotely sensed data such as air-borne radar, they are left with four basic types of digitalgeographic data currently available from the USGS: digitalorthophoto quadrangles (DOQs), digital elevation models(DEMs), digital line graphs (DLGs), or digital rastergraphics (DRGs). The USGS National Mapping Division(NMD) defines these datasets as follows:

DOQ – a digital image of an aerial photograph in whichdisplacements caused by the camera angle and the ter-rain have been removed. They combine the imagecharacteristics of a black and white, color, or colorinfrared photograph with the geometric qualities of aUTM projection map based on the North AmericanDatum of 1983, and have a 1-meter ground resolution.

DEM – a digital record of terrain elevations for groundpositions at regularly spaced intervals primarilyderived from published USGS topographic maps.

DLG – these are spatial representations of planimetricinformation using points, lines, and areas. Currently,the USGS DLG data are distributed as level 3 data(DLG-3), indicating that the data contain a full rangeof attribute codes, have full topological structuring,and have passed extensive quality-control checks.

DRG – a scanned raster image of a USGS topographicmap whose data, inside the map neatline, is georefer-enced to the surface of the earth. The raster data havebeen scanned at either 200 or 250 dots per inch (dpi).

Of these, only the DRG provides the cartographerwith all of the framework themes. Yet, given their poorresolution and low quality, cartographers are hesitant touse this product. As stated above, all four types of digitalbase data must be in some way manipulated and/or com-bined to generate an accurate, clearly legible, and aestheti-cally pleasing base upon which to drape a geologic map.Tables I and II summarize the advantages, disadvantages,and necessary modification requirements for each of thesedatasets.

Very effective base maps can be constructed by com-bining a DOQ or a DEM with a DLG-3, if any or all of itsframework layers are available, but place names and sym-bology must be added. Combining a DOQ or a DEM witha DRG will also provide effective base coverage if thetopographic map data the DRG contains has a suitablecontour interval and the data is cleanly sampled. DLG-3s,again, where available, have the advantage of being updat-ed and manipulated because of their vector format, where-as DRGs ultimately only capture the time in which theoriginal paper map was produced. Unfortunately, due tothe linear nature of DLG-3s, the data often contain artifi-cial straight line segments and “jaggies,” non-smooth,jagged departures from true curves at junctions betweenstraight line segments.

DLG-3 VS. DRG CASE STUDY: PLACITAS,NM 7.5-MINUTE QUADRANGLE

The cartographer, to provide NSDI compliant basedata, must rely on either the DLG-3 or the DRG. In areasof complex topography with high relief and a small con-tour interval, neither digital dataset is adequate. To illus-trate this, I present below a case study of a small area ofhigh relief in the Placitas, NM 7.5-minute quadrangle,underlain by a complex Paleozoic and Proterozoic geolo-gy.

Methods

Transportation, hypsography, hydrography, and PLSSDLG-3 layers were imported from SDTS into Arc/info for-mat. (Only elevation data is found in the study areadepicted below.) In the interest of time, contour labelswere left off. Labeling contours with the proper italic font,point size, and appropriate line break distance is a time-consuming task in Arc/Info. The resulting data wasexported as a TIFF, imported into Adobe Photoshop, con-verted to grayscale, and sampled at both low (72 pixels/in,roughly equivalent to dpi) and relatively high (600

GROUND CLUTTER: SHORTCOMINGS OF THE MODERN DIGITAL TOPOGRAPHIC BASE

Advantages

DOQ DEM DLG-3 DRG• Excellent as a shaded • Excellent as a shaded • Vector data allows the • Inexpensive data

background when draped background when draped user to update map (~$1/quad) by a DLG-3 or a DRG by a DLG-3 or a DRG changes

• Transportation, hydro- • Hillshading feature • Points, lines, and areas • All types of frameworkgraphy, landmarks gives an appearance of have attributes data are present (buildings, etc.) clearly shaded relief visible

• Ability to generate a drainage network

Disadvantages

• No hypsography, political • No cultural data, • Sparse coverage • Data quality poor due to boundaries, PLSS, political boundaries, low resolution capture typography PLSS, typography (200-250 dpi)

• Data modification (merging • Time dependent (static individual layers with their raster data records age attribute data, assigning of the paper map) plot order, line weights, etc.) time consuming and cumbersome

• Original data quality varies, • Significant cartographic dependent on digitization modification necessary quality and quality-control (remove colors, hachured

areas, areas with screen tints

• Data contain artificial straight line segments andjaggies

Table I. Advantages and Disadvantages of Digital Geographic Base Datasets.

DOQ DEM DLG-3 DRG• Spatial Data Transfer • STDS import • STDS import • Run image grid in Arc/Info

Standard (STDS) import and remove colors

• Combine quarter quad data • Project data (if necessary) • Merge each data layer • Generate postscript at and seam together with its attribute set appropriate scale and

convert to grayscale TIFF

• Add hypsography, • Generate hillshade and • Establish plot order, line • Remove additional political boundaries, PLSS, contours, label contours weights, etc., label contours hachured lines, screen typography tints

• Generate drainage net (fill • Convert to grayscale TIFF sinks, create flow direction and flow accumulation grids, stream order and stream line)

• Add political boundaries, • Add typography PLSS, typography

Table II. Necessary Modifications of Digital Geographic Base Datasets.

23


pixels/in) resolution. Areas of green color on the DRGwere rendered white, hachured inundation areas and areasof urban purple and gray screen tints were removed. Theimage was then converted to both bitmap (1 bit data) andgrayscale (16 bit data) and sampled at 72 and 600 pixels/in(Figure 1).

Results

The data depicted in Figure 1 are discouraging. TheDLG-3 data at high resolution gives the greatest clarity butthe linear nature of the data is often obvious and “unnatur-al.” The DRG data clearly shows how black pixels repre-senting lines converge, creating an overall blurring effect,as to render the data essentially useless. The low resolu-tion of the initial 250 dpi scan of the paper map is appar-ent and cannot be fixed, unless done so pixel by pixel inAdobe Photoshop or comparable software.

SO WHAT DO WE DO?

If we are in the business of making geologic maps andare therefore necessarily too busy making geologic mapsto have to additionally make the accompanying digitalbase map, we have three basic options. (1) Nothing/AsLittle As Possible. In this scenario we utilize either theDLG-3 or DRG, perform as little modification as neces-sary and accept the limits of our data. (2) Beg. In thiscase we implore the National Mapping Division of theUSGS to seriously consider the end users, with their mod-ern, powerful, hard drive space-rich computers and re-scanthe >55,000 7.5-minute quadrangles of the U.S. at a reso-lution of at least 600 dpi, a level currently accommodatedby most output devices. Why not ask for the data in sepa-rates as well? (3) Fall Back On Memories Of Pre-digitalCartography. At the New Mexico Bureau of Mines andMineral Resources, we initially order and purchase photo-

Figure 1. Comparison of DLG-3 base data to DRG base data for a small area of the Placitas, NM 7.5-minute quadran-gle depicted at 150% at low and high resolution. Arrows point out several jagged straight line segments common inDLG-3 data.

graphic composite negatives from the USGS-NMD (theseare also used to make greenlines). Contract services areobtained to make photographic contact positive prints (weno longer own a Photo Mechanical Transfer camera, hav-ing surplused it at the onset of the digital age), and furthercontracts are made to generate a single 100% TIFF scan ofthe photographic positive image at the highest resolutionpossible, as 16 bit data. We are currently able to acquirescans locally at 400 dpi. The scanned raster data is then

“rubber-sheeted” about the 16 standard grid control pointsin Arc/Info to remove distortion, and transported to AdobePhotoshop. The image quality can then be significantlyimproved by using sharpening filters and adjusting spec-trum levels prior to conversion to bitmap. In such a man-ner we create our own DRG data, at a total cost of $120for the negative, the film work, and the scan. As seen inFigure 2, the quality of the data is far superior to theUSGS 250 dpi DRG data, even at only 400 dpi.

GROUND CLUTTER: SHORTCOMINGS OF THE MODERN DIGITAL TOPOGRAPHIC BASE

Figure 2. Scanned TIFF image of contact print made from photographic composite negative obtainedfrom USGS-NMD (image at 200% size compared to those of Figure 1). Contact print was scanned at 400dpi and sampled at 600 pixels per inch.

25

27

INTRODUCTION

As the U.S. Geological Survey’s (USGS) NationalMapping Division (NMD) is challenged to do more withless and at the same time increase its response to customerconcerns, the traditional methods of revising the presentprimary series of topographic maps are no longer feasible.(Primary series maps include 1:20,000-scale quadranglesof Puerto Rico, 1:24,000- or 1:25,000-scale quadrangles ofthe conterminous United States and Hawaii, and 1:63,360-scale quadrangles of Alaska.) The NMD is responding bycontinually experimenting and devising new methods bywhich graphic revision can be accomplished. This pro-posed approach will provide the means to determine anduse the most cost-efficient method(s) for graphic revision.By devising such a revision methodology, that encompass-es the concerns expressed in this paper and responding tothe needs of our graphic customers, the NMD will becomemore responsive to the graphic segment of the consumerbase and in turn minimize the cost of production.

CHANGES IN THE GRAPHIC REVISIONPROGRAM

The USGS National Mapping Program (NMP) hasbeen in existence for over 100 years, and during that timethe USGS has established itself as one of the leading civil-ian mapping agencies in the world. The role of the NMPhas been to provide accurate, detailed cartographic andgeospatial information for the Nation. This mission hasnot changed; however, the means of accomplishing thatmission and the ways of communicating the resultinginformation have changed dramatically. During its exis-

tence, the USGS has established a consumer base withrequirements that greatly exceed the original mandate ofsupporting just the Federal mapping community. Recentconcerns from consumers have heightened our awarenessof the effects on mapping that the NMD has had outsidethe Federal community. They are prompting us to look fora more responsive and inexpensive map maintenance pro-gram for our traditional products.

Although the USGS-NMD has historically beenknown for the large-scale topographic maps producedthrough its Graphic Revision Program, it has, along withthe rest of the mapping industry, made the transition tocomputer-aided mapping. (Information on NMD’sGraphic Revision Program can be found on the Internet at<http://mapping.usgs.gov/mac/isb/pubs/factsheets/fs01698.html>.) During the late 1970’s and early 1980’s, NMD’sVector Program was created. This program initiallyfocused on the production of digital vector data that usedthe graphic map as a source. The data collected represent-ed the various feature categories shown on the map. Thesecategories included transportation, hydrography, hypsogra-phy, and boundaries, among others. The growth anddevelopment of these digital data, defined as digital linegraphs (DLG), along with the 1990 Census Program,became the foundation for the geographic information sys-tem (GIS) industry and the focus of our future. Our VectorProgram contributed significantly to the development ofconcepts that led to the 1994 signing of PresidentialExecutive Order (EO) #12906 establishing the NationalSpatial Data Infrastructure (NSDI) and to the introductionof the framework concept. (Information on NMD’s VectorProgram is available on the Internet at<http://mapping.usgs.gov/dpi/vector8.html>.) Althoughthis EO was a consolidated effort between multiple Federal

The Evolution of Topographic Mapping in theU.S. Geological Survey’s National Mapping Program

By Robert M. Lemen

U.S. Geological SurveyNational Mapping Division

Mid-Continent Mapping Center 1400 Independence RoadRolla, Missouri 65401

Telephone: (573) 308-3736Fax: (573) 308-3652



agencies, the foundation for it was based on the work pre-viously done by the NMD.

Throughout the evolution of the Vector Program, thegraphic customer was viewed as making the digital transi-tion along with the rest of the industry. It was assumedthat since graphics were the source for the Vector Program,graphics could easily be derived from this digital product.Digital revision would support the graphic revisionrequirement since the vectors would reflect change and, inturn, serve as sources for the new graphics. As this con-cept matured, graphics became more and more dependentupon the existence of digital vectors, to the extent thatboth the Vector and Graphic Revision Programs wereentirely dependent upon the existence and collection ofvector information. The NMD still retained an analoggraphic process for small-scale and special products, butour primary series became dependent upon the vectorsource. This dependency has become the underlying focusof many of our present digital production policies andstandards. (NMP Geospatial Standards are available onthe Internet at <http://mapping.usgs.gov/standards/>.)

In an effort to reduce the expense of both digital andgraphic products and to increase their availability, theNMD looked at the product content. In an effort toimprove production efficiencies, the NMD reviewed andmodified standards for the Vector Program that affectedthe content of the conventional USGS graphic products. (Itshould be noted that many content changes also cameabout because the NMD no longer had the resources toperform field verification.) The public expressed concernsthat these vector policies would have negative impacts onthe quality and availability of future USGS topographicmaps. Many customers have encouraged us to revisitsome of these decisions.

Digital production has traditionally been cooperatordriven and focused on category-specific collection andrevision. This leaves those unrevised or uncollected mapcategories out-of-date when compared to the newer vectordata. This becomes a significant problem when trying toproduce a consistent graphic product. The vector-basedproduct generation process uses digital vector data to pro-duce a graphic map. It requires the update of map featuresto revise the entire quadrangle. Graphic customers haveexpressed concern that they do not want to be burdenedwith the additional expense of paying for the production ofvectors to produce the graphic product they want. Aprocess that integrates both color separates and existingvector data provides a less expensive revision alternative.

As the GIS industry matures and the State and localgovernments move their daily activities to GIS-basedapplications, the NMD is going to be challenged to main-tain continuity with regard to the accuracy and content of

local GIS holdings. The data sets held and maintained byother agencies in many cases show additional content andare of a newer vintage. To republish a graphic that is outof date compared with local data is embarrassing for theUSGS. The Graphic Revision Program needs to maintainconsistency with new local data to avoid this problem.This becomes increasingly difficult if policy requires datacompatibility between the Vector and Graphic RevisionPrograms.

The NMD does not have the resources and capacity tomeet the demands of the Graphic Revision Program.Therefore, to ensure the production and availability ofrevised graphic maps, we must use data sources producedby other organizations. A major component of the NSDI isthe development and implementation of a national geospa-tial data framework. As part of the framework concept,NMD will pursue partnerships with other Governmentagencies and the private sector to use locally produceddata that can contribute to its mapping program. We fore-see spatial data produced by public and private sectororganizations contributing to the revision of our maps. Byimplementing integration techniques, we can take advan-tage of these local up-to-date data holdings.

The NMD recognizes the need to find commonground between digital and graphic customers at a mini-mal expense and without degrading its products. A keyapproach to this problem is to separate the product depen-dencies and focus strictly on the customer’s needs. Thisapproach prevents the vector cooperator from being bur-dened by the expense of supporting a graphic product. Atthe same time, when applied to the Graphic RevisionProgram, it provides a significant cost savings to thegraphic customer. If the Graphic Revision Program is notburdened by the expense to complete vector data to anachievable vector standard, the overall expense to thegraphic customer is significantly less.

A goal of the Graphic Revision Program is to producestandard cartographic products regardless of the datasources. The focus needs to be broad enough in scope thatit ranges from graphic revision using a full range of digitalsources to a reprint process using existing analog tech-niques. Vectors should be used if readily available butshould not be required. If vectors are collected andrevised during the graphic process, they should be provid-ed to the Vector Program as transactional updates. If acustomer desires both vectors and graphics, the VectorProgram can deliver a geographic source for use in anautomated generalization process to generate graphics.

The NMD has developed a raster/vector process thatwill work with existing vector and raster capabilities. Thisprocess allows the revision of the graphic map from exist-ing vector or graphic source data. This approach ensures a

diverse graphic revision process that uses both analog anddigital sources to update products. Figures 1 and 2 illus-trate the vector and raster revision processes.

NMD’S CURRENT REVISION PROCESS

The NMD’s current method of revising most of its pri-mary series graphic maps is called the raster graphic revi-sion (RGR) process. This method has proven to be themost cost-efficient process for those users who are onlyinterested in acquiring a revised graphic product. The ini-tial prototype project was completed 1.5 years ago. Sincethe completion of this project, the process has gonethrough refinements and enhancements and is the methodused for most of NMD’s in-house graphic revision.

The RGR process was developed to reduce the pro-duction costs and the amount of time it takes to revisequadrangle maps, while still meeting customers’ needs fora graphic product. This method starts by scanning the mapseparates at 1,000 dots per inch (dpi) and then making theappropriate additions and deletions to the map detail onthe basis of new data sources. For the most part, thesources used are digital orthophoto quadrangles (DOQ).(Information on DOQ’s is available on the Internet at<http://nsdi.usgs.gov/products/doq.html>.)

There are two categories of map revision; completerevision and basic revision. In the complete revisionprocess, all features are corrected and updated. Content isvalidated by field checking against ground truth. Contoursare revised. The revised map meets all current NMD stan-dards for feature content and National Map AccuracyStandards for positional accuracy.

In the basic revision process, many features arerevised by interpreting image sources, such as digitalorthophotos or aerial photographs. The features are notverified through field checking, and contours are generally

not revised. The revised map maintains the positionalaccuracy of the previously published map. Most resourcesare focused on basic revision because this method is lessexpensive and more maps can be revised in a given periodof time.

As part of the NMD’s revision program, a new digitalraster graphic (DRG) will be produced for every revisedgraphic. Recently DRG’s, scanned raster images of USGStopographic maps, were completed for all of the USGSprimary series maps. (The paper 7.5-minute topographicmaps were scanned at 250 dpi.) Since they were producedfrom the existing paper topographic maps, the DRG’s areonly as current as the map sources. Now with the rasterrevision process in place, the NMD can produce a newDRG that reflects the revised graphic. (Information aboutNMD’s DRG’s is available on the Internet at <http://mcm-cweb.er.usgs.gov/drg/>.)

Although the new DRG’s are being produced througha different process, they will adhere to the same standardsand specifications that were used during the production ofthe initial DRG’s. Original DRG coverage for the countrywas produced by scanning the paper topographic mapitself. During the raster revision process, each map sepa-rate that is scanned and then revised through the raster editprocess will be used to produce the DRG. Where wemade one scan of the paper map, we will now have araster scan of each revised color separate. These rasterscans will be combined to produce the DRG.

Although this paper describes the transition of NMD’smethodology for revising its primary series maps, it shouldbe noted that the NMD does not rely solely on in-houseproduction to accomplish its goals. A significant part ofthe map production will be accomplished by the privatesector through contract work. Within the past year, theNMD has begun contracting some of its revision work.The contractor will steadily increase its share of the revi-sion program over the next few years.

THE EVOLUTION OF TOPOGRAPHIC MAPPING IN THE USGS’S NATIONAL MAPPING PROGRAM

Figure 1. Diagram illustrating graphic map revisionthrough the vector process.

Figure 2. Diagram illustrating graphic map revisionthrough the raster process.

29


CONCLUSION

As the Nation’s need for mapping information contin-ues to grow, the NMD will address this need and beresponsive to the map users and their requirements. Itsgoal is to develop a long-term maintenance strategy thatmeets the need for revising the primary maps. In additionto providing the user with revised maps more quicklythrough a more cost-efficient method, the NMD wants to

coordinate issues of content, accuracy, and standards withits partners. In NMD, this will be accomplished in greatpart by building cooperative partnerships with its mapusers. These partnerships, through shared costs andresources with other organizations, will enable the USGSto meet the increased demand for revised topographicmaps of the Nation. With the limited resources and dollarsavailable, we are seeking and encouraging such partner-ships to fulfill this need.

The Geologic Mapping Act of 1992 and its reautho-rization in 1997 (PL105-36) requires that a NationalGeologic Map Database (NGMDB) be designed and builtby the U.S. Geological Survey (USGS), with the assistanceof the state geological surveys and other entities participat-ing in the National Cooperative Geologic MappingProgram. The Act notes that the NGMDB is intended toserve as a “national archive” of geologic maps, to providethe information needed to address various societal issues.The Act required the NGMDB to also include the follow-ing related map themes: geophysics, geochemistry, paleon-tology, and geochronology. In this progress report, theterm “geoscience” is used to refer to these five mapthemes.

In mid-1995, the general stipulations in the Act wereaddressed in the proposed design and implementation plandeveloped within the USGS and the Association ofAmerican State Geologists (AASG). This plan was sum-marized in Soller and Berg (1995). Because many mapsare not yet in digital form and because many organizationsproduce and distribute geologic maps, it was decided todevelop the NGMDB in several phases. The first twophases are addressed here. The first and most fundamentalphase is a comprehensive, searchable catalog of all geo-science maps in the United States, in either paper or digital

format. The users, upon searching the NGMDB catalogand identifying the map(s) they need, are linked to theappropriate organization for further information about howto procure the map. (The organization could be a partici-pating state or federal agency, association, or private com-pany.) The map catalog is presently supported by twodatabases developed under the NGMDB project: 1)GEOLEX, a searchable geologic names lexicon; and 2)Geologic Mapping in Progress, which provides informa-tion on current mapping projects, prior to inclusion of theirproducts in the map catalog. The second phase of the pro-ject focuses on public access to digital geoscience maps,and on the development of digital map standards andguidelines needed to improve the utility of those digitalmaps.

In late 1995, work began on phase one. The forma-tion of several Standards Working Groups in mid-1996 ini-tiated work on phase two. Progress was summarized inSoller and Berg (1997 and 1998). At the Digital MappingTechniques ‘98 and ‘99 workshops, a series of presenta-tions and discussion sessions provided updates on theNGMDB and, specifically, on the activities of theStandards Working Groups. This report summarizesprogress since mid-1998. Further and more current infor-mation may be found at the NGMDB project-information

31

The National Geologic Map Database—A Progress Report

By David R. Soller1 and Thomas M. Berg2

1U.S. Geological Survey908 National CenterReston, VA 20192

Telephone: (703) 648-6907Fax: (703) 648-6937


2Ohio Geological Survey4383 Fountain Square Dr.

Columbus, OH 43224Telephone: (614) 265-6988


Web site, at <http://ncgmp.usgs.gov/ngmdbproject>. Thesearchable database is available at<http://ngmdb.usgs.gov>.

PHASE ONE

The Map Catalog

The catalog now contains bibliographic informationfor nearly all formal series USGS maps, USGS maps con-tained in book publications, and maps from the USGSopen-file series. The catalog is estimated to be about 50%complete, and contains georeferenced information forabout 90% of all USGS maps (about 18,500 in the catalog)and 6% of state geological survey maps. This representsmore than a six-fold increase in information since lastyear. Through development of a Web-based data-entryform, and reformatting and enhancements to informationcontained in existing State publications listings, we areworking with nine state geological surveys (Illinois,Indiana, Louisiana, Ohio, Oklahoma, Oregon, Vermont,West Virginia, and Wyoming) and one University(Stanford) to bring their map information into the catalog.Entry of state geological survey publications is now a toppriority.

Geologic Names Lexicon

In April, 1998, an on-line, geologic-names lexicon,GEOLEX, became available at the NGMDB Web site.This lexicon is under construction, and is estimated to beabout 75% complete. At present, GEOLEX contains 90%of the geologic names found in the most recent listing ofUSGS-approved geologic names; this listing was pub-lished in 1996 as USGS Digital Data Series DDS-6, revi-sion 3. Prior to loading into GEOLEX, the information onDDS-6 was consolidated, revised, and error-corrected.Much of the remaining work needed to completeGEOLEX will focus on resolving name conflicts, addingreference summary and other information for each entry,and incorporating geologic names not found on DDS-6 butrecorded in the geologic names card catalog at USGSHeadquarters. GEOLEX is intended to be the comprehen-sive, authoritative listing of geologic names approved forusage by the USGS, and is available as a resource for geo-logic mappers nationwide. Many state geological surveyshave been registering new geologic names with the USGSfor decades, and are encouraged to continue underGEOLEX, through a Web-based application form.

Geologic Mapping in Progress Database

To provide users with information about current map-ping activities at 1:24,000- and 1:100,000-scale (1:63,360-

and 1:250,000-scale in Alaska), a Geologic Mapping inProgress Database has been developed. In 1998, its scopeand design were proposed to the AASG and a prototypedatabase was built. The database has been expanded tocontain information about 1998 mapping activities, and aWeb interface to the database has been designed and built.It is available through the NGMDB home page,<http://ngmdb.usgs.gov>.

Related Databases

Work has begun on a National Paleontologic Databaseand a set of Web pages to permit searches and to providegeneral-interest information in support of the Database. Inthe coming year, efforts will be made to provide links fromthe NGMDB to related databases under construction (forexample, for geophysical and geochemical information).

PHASE TWO

Most efforts related to phase two have been directedtoward the development of standards and guidelines need-ed to help National Cooperative Geologic MappingProgram participants more efficiently produce digital geo-logic maps, and to produce those maps in a more standard-ized and common format among the various map-produc-ing agencies. Significant progress has been made towarddeveloping some of these standards and guidelines, andincreased efforts will be devoted in the coming yeartoward improving public access to digital maps.

Standards Development

The following summaries concern activities of theAASG/USGS Standards Working Groups formed in mid-1996. General information about the Working Groups, anddetails of their activities, are available at<http://ncgmp.usgs.gov/ngmdbproject>.

Geologic Map Symbols

A draft standard for geologic map line and point sym-bology and map patterns and colors, published in a USGSOpen-File Report in 1995, was revised by the NGMDBproject team and members of the USGS Western RegionPublications Group and was circulated for internal reviewin late 1997. The revised draft is now being prepared as aproposed Federal standard, for consideration by theFederal Geographic Data Committee (FGDC). We antici-pate that the draft will be submitted to the FGDC in 1999;informal USGS release of the map symbols and patterns inPostscript format also is anticipated in 1999. To makethese symbols more widely available to the GIS communi-ty, we have negotiated with Environmental SystemsResearch Institute, Inc., a cooperative plan to develop an


Arc/Info and ArcView version of the symbols and patterns;for more information, see<http://ncgmp.usgs.gov/ngmdbproject/standards/carto/car-tomain.html>.

Digital Mapping

The Data Capture Working Group has coordinatedthree annual “Digital Mapping Techniques” workshops forstate, federal, and Canadian geologists, cartographers, andmanagers. These meetings have been highly successful,and have resulted in adoption within agencies of new,more efficient techniques for digital map preparation,analysis, and production. The most recent workshop, heldin Madison, Wisconsin, and hosted by the

Wisconsin Geological and Natural History Survey,was attended by representatives of 42 state, federal, andCanadian agencies and private companies. The workshopproceedings are published (Soller, 1997, 1998, and thisvolume) and served on-line(<http://ncgmp.usgs.gov/pubs/of97-269>;<http://pubs.usgs.gov/openfile/of98-487>; and<http://pubs.usgs.gov/openfile/of99-386>). Copies of theProceedings may be obtained from Soller or Berg.

Map Publication Requirements

Through the USGS Geologic Division InformationCouncil, one of us (Soller) led development of the USGSpolicy “Publication Requirements for Digital MapProducts” (enacted May 24, 1999). A less USGS-specificversion of this document has been developed by the DataInformation Exchange Working Group and presented fortechnical review at a special session during the DigitalMapping Techniques ‘99 workshop (this volume). Therevised document (entitled “Proposed Guidelines forInclusion of Digital Map Products in the NationalGeologic Map Database”) has been submitted to theAASG Digital Geologic Mapping Committee for consider-ation as a guideline for newly-produced maps availablethrough the NGMDB.

Metadata

The Metadata Working Group developed its finalreport in 1998. The report provides guidance on the cre-ation and management of well-structured formal metadatafor digital maps (see <http:// ncgmp.usgs.gov/ngmdbpro-ject/standards/metadata/metaWG.html>). The report con-tains links to metadata-creation tools and general discus-sions of metadata concepts (see, for example, the metada-ta-creation tools, “Metadata in Plain Language” and otherhelpful information at <http://geology.usgs.gov/tools/meta-data/>.

Geologic Map Data Model

Following numerous presentations, discussions, andprogress reports (for example, Raines and others, 1997), inOctober, 1997, the Working Group posted to the NGMDBproject’s Web site for public comment a report describingthe proposed standard data model. To permit thoroughevaluation of the data model, the concepts and specifica-tions described in that report must be translated into soft-ware tools that 1) organize and manage the geologic mapinformation in the data-model format, and 2) offer a user-friendly interface for data entry and analysis. TheWorking Group began building some prototype tools inmid-1997. Through presentations and discussions withpotential users, the tools were refined.

The data model and tools were presented for discus-sion in a special session at the Digital Mapping Techniques’98 workshop in late May, 1998. A number of attendeesfrom the state geological surveys expressed interest inhelping to refine the model and develop it into a geo-science-community standard. It was agreed that anextended period of evaluation among the states and USGSwould begin as soon as possible.

At a three-day workshop in June, 1998, the first for-mal review of the data model was conducted. The work-shop was attended by 28 members of the USGS, state geo-logical surveys, and the federal and provincial surveys ofCanada. An overview of the conceptual model and soft-ware tools was followed by a hands-on session and facili-tated discussion. The workshop conclusions were to: 1)proceed with development of the data model and softwaretools; 2) broaden the participation in its further develop-ment; 3) begin the extended period of on-site evaluation atstate, provincial, and federal surveys; and 4) work towardadoption of the data model as a national standard.

To facilitate the review and discussion of the datamodel after the June workshop, a Web conference wasdeveloped (see <http://geology.usgs.gov/dm/>).Discussion topics include: 1) general conceptual issuesrelated to the data model and geologic mapping; 2) specif-ic problems with the data model; 3) development of stan-dard geologic terms; and 4) software tools. This site, andthe NGMDB project’s Web site, also offer access to thedata model report and software tools.

The workshop’s technical review of the data modelsignaled completion of the Data Model Working Group’stask. In late 1998, a mechanism to coordinate the contin-ued development of the data model was proposed to theUSGS, AASG, and the Geological Survey of Canada.This mechanism, a Data Model Steering Committee, wasaccepted in early 1999 and the first meeting was held soonthereafter at the USGS National Center. The SteeringCommittee is now forming numerous Technical Teams toaddress discrete tasks deemed necessary to further devel-opment and adoption of the data model. These TechnicalTeams will address issues such as: the inclusion of stan-

THE NATIONAL GEOLOGIC MAP DATABASE: A PROGRESS REPORT 33

dard scientific terminology in the model; refinement of theconceptual data model; and development of software tools.Before the Technical Teams begin work, the SteeringCommittee will contact various users of digital geologicmaps and conduct a requirements analysis, to provideinformation on how they use, or would like to use, digitalmaps; this information then will help guide the activitiesof the Technical Teams. Information about the SteeringCommittee and Technical Teams can be found at<http://geology.usgs.gov/dm/>.

REFERENCES

Raines, G.L., Brodaric, Boyan, and Johnson, B.R., 1997,Progress report—Digital geologic map data model, in DavidR. Soller, ed., Proceedings of a workshop on digital map-ping techniques: Methods for geologic map data capture,

management, and publication: U.S. Geological SurveyOpen-File Report 97-269, p. 43-46,<http://ncgmp.usgs.gov/pubs/of97-269/raines.html>.

Soller, D.R., editor, 1997, Proceedings of a workshop on digitalmapping techniques: Methods for geologic map data cap-ture, management, and publication: U.S. Geological SurveyOpen-File Report 97-269, 120 p.<http://ncgmp.usgs.gov/pubs/of97-269/>.

Soller, D.R., editor, 1998, Digital Mapping Techniques ‘98 —Workshop Proceedings: U.S. Geological Survey Open-FileReport 98-487, 134 p. <http://pubs.usgs.gov/openfile/of98-487/>.

Soller, D.R., and Berg, T.M., 1995, Developing the NationalGeologic Map Database: Geotimes, v. 40, no. 6, p. 16-18.

Soller, D.R., and Berg. T.M., 1997, The National Geologic MapDatabase—A progress report: Geotimes, v. 42, no. 12, p.29-31.


INTRODUCTION

The enabling legislation for the National GeologicMapping Act of 1992 includes requirements for develop-ment of a National Geologic Map Database (NGMDB),and for standards and guidelines necessary to support itsready access and use by the public. [See the 1992 Act andits 1997 Reauthorization at <http://ncgmp.usgs.gov>]. Asdetailed elsewhere (for example, in Soller and Berg, thisvolume), the National Geologic Map Database (NGMDB)is designed as a distributed system developed through con-sensus among its primary builders, the Association ofAmerican State Geologists (AASG) and the U.S.Geological Survey (USGS). It is not intended to be a sin-gle entity managed by one group. Because of the rapidevolution of digital mapping and methods for distributingmap products, the NGMDB is not a conventional database;primary emphasis has been to develop a searchable catalogof available maps in paper and digital form, at<http://ngmdb.usgs.gov>, and a set of standards and guide-lines that are agreeable and beneficial to each agency,described at <http://ncgmp.usgs.gov/ngmdbproject)>.

At some time in the future, the system we are buildingwill lead to a distributed database of digital maps. Thesemaps will be managed by each producing agency and

served to the public in a fashion that will enable users toaccess maps from disparate agencies and use them togeth-er, for analysis and display, without the current level ofeffort required to integrate them. We are now, as a geo-science community, discussing the level of standardizationamong agencies that is appropriate (and affordable). Thispaper is intended to contribute to that process.

In August, 1996, technical representatives from theAASG and the USGS met in St. Louis, MO, to identify thestandards and guidelines needed to support the NGMDB.As a result, several AASG/USGS Working Groups wereformed to conduct the needed work. [Informal minutes ofthe meeting are available at<http://ncgmp.usgs.gov/ngmdbproject/standards/mtgs/St.Louis>.]

The Data Information Exchange Working Group wascharged to develop certain proposed guidelines for produc-ers of digital geologic maps, specifically: “what files andfile structure should be “packaged” into a digital geologicmap, to promote useability?” In recent years, certainlywithin the past five, the USGS and state geological sur-veys have increasingly produced their maps in digital for-mat, and provided them to the public through variousmechanisms including traditional over-the-counter salesand the Internet. This Working Group was formed to pro-

35

Proposed Guidelines for Inclusion of Digital Map Productsin the National Geologic Map Database

By David R. Soller1, Ian Duncan2, Gene Ellis3, Jim Giglierano4, and Ron Hess5

1U.S. Geological Survey908 National CenterReston, VA 20192

Telephone: (703) 648-6907Fax: (703) 648-6937


2Virginia Division of Mineral Resourcese-mail: [email protected]

U.S. Geological Surveye-mail: [email protected]

4Iowa Geological Surveye-mail: [email protected]

5Nevada Bureau of Mines and Geologye-mail: [email protected]

vide technical advice that could lead to more uniformity tothe manner in which geologic map information is providedin these products; not the scientific content, but mostly inthe organization and content of the various files thataccompany the map. Further information on this WorkingGroup can be found at <http://ncgmp.usgs.gov/ngmdbpro-ject/standards/dataexch/dataexchWG.html>.

This paper is authored by the AASG/USGS DataInformation Exchange Working Group, and contains itsreport. The report is based on earlier (October, 1996) rec-ommendations to USGS participants in the NGMDB andon a new USGS Geologic Division policy (“PublicationRequirements for Digital Geologic Map Publication”,enacted May, 1999) for which one of us (Soller) was partlyresponsible while a member of the USGS GeologicDivision Information Council. To gather further technicalinput, the proposed guidelines were presented at theDigital Mapping Techniques ‘99 meeting, in a special dis-cussion session. The document was well received, andminor changes were suggested. The revised report (con-tained herein) recently was submitted to the AASG’sDigital Geologic Mapping Committee for consideration asa voluntary guideline applicable to all participants in theNGMDB. We emphasize that this is a guidance document,developed collaboratively, and is not meant to proscribehow an agency must release its data. It is intended to pro-vide mapping agencies with information on helpful, typicalmap data files and documentation that commonly areincluded in a digital map product made available to thepublic.

REPORT TO THE AASG DIGITALGEOLOGIC MAPPING COMMITTEE

Proposed Guidelines for Inclusion of DigitalMap Products in the National Geologic MapDatabase

NOTE - the following DRAFT document concerns apotential cooperative agreement between the AASG andthe USGS, in support of requirements of the NationalGeologic Mapping Act. The Act stipulates the develop-ment of various standards, to support the NationalGeologic Map Database. This document addresses thegeneral format of map products to be made availablethrough the Database. It does not include discussion of astandard data model. It is meant to be an informativeguideline, not a requirement, and is intended to providethese agencies and the public with more compatible, betterdocumented, and hence more useable map products.Because digital mapping has evolved rapidly, the specifi-

cations in this guideline periodically will be revisited, andmay be revised.

Geologic map information supports the needs of abroad range of users. To increase its utility and to promoteintegration with related data sets produced by other organi-zations, the information should be readily available, well-documented, and well-structured. The National GeologicMapping Act of 1992 and 1997 articulates these goals, bystipulating development of various standards and guide-lines necessary to promote the more efficient use and shar-ing of information. The Act calls for these standards to bedeveloped by the AASG and the USGS in support of theircooperatively-built national resource, the NationalGeologic Map Database (NGMDB). Information is avail-able about the standards now under development, at<http://ncgmp.usgs.gov/ngmdbproject>.

Map users are, increasingly, integrating in a GIS themap products of various geological surveys and other mapproducers. This integration is made easier where the prod-ucts are well-documented and share certain common ele-ments (e.g., metadata, browse graphics, readme files).This document addresses only these general elements of amap product, and is intended to promote uniformity amongthe agencies that collaborate to build the NGMDB. It doesnot include requirements for a standard geologic map datamodel, nor does it stipulate scientific content or data inter-change format. Those more complex elements of a geolog-ic map and its presentation will require far more discus-sion among the AASG and USGS.

In the transition from production of maps solely onpaper, to production of maps in both digital and paper for-mat, the map’s geographic, cartographic,and scientificinformation has been transformed from a strictly visualmedium to one based on electronic files. “Digital” mapsnow commonly contain the coordinates for various mapfeatures, and a database of information about the features,which users may analyze. This document addresses therequirements for preparing a single digital map product forpublication, and does not address the integration of dataacross maps of adjacent areas.

1. CONFORMANCE TO EXISTING REQUIRE-MENTS — Digital map products (referred to as “prod-ucts”, below) included in the NGMDB will conform to therespective agency’s policies and guidelines for approvaland publication of products. For example, USGS mapproducts contributed to the NGMDB will conform toDivision and Bureau policies, including the requirementsof Executive Order 12906, USGS Manual chapter 504.1,and Geologic Division Policy Manual chapter 6.1.3.

2. SCOPE AND RESPONSIBILITIES — Theserequirements apply to products intended for release to thepublic in both formal and open-file series. Each agency is


responsible for promoting conformance of their productsto these guidelines.

3. DATA FORMATS — A specific data format is notrequired, because of the variety of data systems employedby all cooperators in the NGMDB, and because theNGMDB does not yet provide an online mechanism forusers to display and query map data from various agencies.Agencies are, however, urged to provide their map prod-ucts in one or more commonly-used data formats (forexample, Arc/Info export and/or Shape format, AutoCADformat). If the map data are expressed in a non-propri-etary format that is not supported by published documenta-tion, the format should be fully and clearly documented inthe product.

4. ASSOCIATED FILES — All associated files, tab-ular and otherwise, containing attribute data should accom-pany the map data files. Lookup tables and color and linepalettes (e.g., Arc/Info symbolsets and shadesets) alsoshould be provided to permit users to display the map datainteractively to a monitor.

5. FILE NAMING CONVENTION — For thewidest possible usage, file names should conform to the“8.3” convention. This convention requires that file namesbe limited to 8 characters or less, followed, if needed, by aperiod and a 3-character extension. An example would bethe file name “readme” or “readme.txt”. The name andextension should be entirely composed of lower-case (notmixed-case) letters, numerals, underscore, and hyphen.The name should begin with a letter.

6. COORDINATE SYSTEMS — Map data providedin geographic coordinates (latitude and longitude) is mostgenerally useable. The author may choose to provide themap data in geographic coordinates and/or in projectedcoordinates, in the map projection and ground units typi-cally used for maps of that scale and location (e.g., theUTM projection for 30-minute by 60-minute,1:100,000-scale quadrangle maps, with ground units in meters). Toavoid loss of data quality due to resampling during projec-tion, raster thematic (e.g., maps showing spatial variationof a single phenomenon, such as geophysical data) shouldat least be provided in the original, unprojected form. Ifthe GIS software does not create a file containing essentialinformation about the projection, such a file should be cre-ated by the author.

7. BASE MAP — Wherever possible, map productsshould be georeferenced to a digital base, preferably theone on which the map was compiled. As a service tousers, the author may elect to include the base map withthe product; this is highly recommended if the base is not

published or is not commonly available. If a digital basewas used, and if the base was revised to correct for spatialor attribution errors, it should be supplied (in vector orraster format) with the product. Revisions to publishedbase maps should be supported with metadata thatdescribes the data processing. However, not all geologicmaps are compiled on a digital base, generally because oneis not available. In such cases, it is suggested that a) thebase be scanned and georeferenced, b) the geologic mapbe geoferenced to the base, and c) the base be provided, invector or raster format, with the product.

8. METADATA — All geologic and base map datashould be documented with metadata conforming to theContent Standard for Digital Geospatial Metadata(CSDGM) of the Federal Geographic Data Committee(FGDC). Conformance of the metadata to the structuredefined in the CSDGM can be determined using the USGSmetadata parser “mp”. This parser verifies the specificindented-text format compatible with the Geospatial DataClearinghouse. This specification should not, however,preclude each agency from exploring other options formanaging metadata, including relational databases.

9. README FILE — A brief, overall introductionand guide to the product should be included in a plain-textfile named “readme” or “readme.txt”. This file shouldinclude, but is not limited to, the identity of the product, abrief product description, introductory instructions on howto extract information from the product file(s), a table ofcontents describing how the product’s directories and filesare organized, and the location of the detailed metadata.

10. BROWSE GRAPHIC — A low-resolution“browse” graphics file that represents the finished mapproduct should be provided in GIF, JPEG, TIFF, EPS orPDF format. This file is intended to be a relatively simpledepiction of the data that enables the user to quickly visu-alize the map from the author’s perspective. Typically,this graphics file is not a fully-detailed depiction of themap data; in such cases the graphic should contain, next tothe map image, the following disclaimer: “NOTE: Thisimage is not an authoritative representation of the data.”

11. PLOT FILE — The author is encouraged to alsoinclude a “plot file” (preferably EPS or PDF), intended toprovide the user with the author’s full interpretation of themap data. Commonly, these plot files are as detailed aspublished geologic maps. The decision to include a plotfile might be based on the map content and complexity (isthe product a complex, multi-purpose geologic map, orsimply a derivative map showing areas of greater and less-er geologic hazard?) and the size of the file (will it, withthe map product, fit on the intended media?). If a plot file

PROPOSED GUIDELINES FOR DIGITAL MAP PRODUCTS IN THE NATIONAL GEOLOGIC MAP DATABASE 37

is included, the author should note, in the metadata orreadme file, the plotter and the RIP software with whichthe map has successfully been plotted, and the dimensionsof the plot image.

12. PRODUCT FILE — The product should bepackaged in one or more files in a universal, cross-plat-form format. At present, the “tar” and “gzip” formats bestfit this description. The decision of whether to use oneproduct file or more should be based on the content andsize of the product. Generally, one product file is pre-ferred because product integrity is more easily maintained.However, if the product is relatively large and contains anextensive base map and/or a large plot file, the author maychoose to package the plot file or base map in a productfile separate from the geologic data. In that case, both

product files would contain the readme file. The authorshould provide in the readme file the information neededto unpack the product file; this may include providingURLs where tar or gzip software may be freely obtained.The product file is intended to provide users with a simplemeans for copying the product to a local disk, which isespecially helpful for products with many data files.

13. SUPPORTING DOCUMENTATION —Potential users of the data may want a brief overview ofthe product before deciding whether to acquire it.Therefore, authors should provide the following separatefiles to accompany the single-archive file containing theproduct (these are duplicates of files contained in the prod-uct): the readme file, the browse graphic, and the metadatafile in plain-text and, optionally, in HTML format.


39

Geological surveys have begun to produce maps usinggeographic information systems (GIS) in response to therecent growth in the use of this technology by decision-makers and users in general. In the past, printed reportsand maps conveyed geologic information. Now, a surveyorganization can deliver the same geologic informationseveral ways: as a printed report and map, digital imagefiles, and GIS data files.

Authorship and citation of paper products in literaturefollow long-standing and well-established conventions, butwe have not found any acceptable or established conven-tion for digital publications. This paper reviews severaldigital products, considers the relationship of authors totheir work, and proposes authorship and citation guidelinesfor these new digital products.

BACKGROUND

Geoscientists are rightly concerned about retainingauthorship for map products that they have created.Authors of published maps should expect to see their nameon every paper or raster image of a geologic map wherethey were responsible for the collection of geologic data,synthesis of the data, and “assembly” of the original map.However, authors of these maps should not be liable forerrors in a digital reproduction produced by someone else.Vector-data authors should have their name associated withdigital files that they create because they share partialresponsibility with the original map author for accuracyand error of data in those files. Interestingly, users of the

digital files in another GIS or computer application maynever look at the original map image (printed or raster)when they query and extract selected point, line or areatable data. Clearly, we need a consensus on equitable andclear guidelines for establishing credit and culpability fordigital (GIS) data.

The issue is further complicated because the term“digital” map has been applied to a product derived fromany of three different procedures. A “digital geologicmap” may refer to a simple copy created by scanning anoriginal paper map and producing a raster image. Second,a scanned map image file may be enhanced using comput-er-graphics software (Adobe, Corel, etc.) to improve colorsand linework, to add text, or to alter the explanation (leg-end). Third, a publication-quality “digital” geologic mapmay be assembled by capturing the map data in the numer-ous spatial data (vector) files of a GIS system followed bymanipulation of these files to produce a map image. Here,the “intermediate” product (vector files) are also a desiredend product.

With each of these three different procedures, there isa correspondingly different level of effort and technicalcompetence required by the creator of the digital product.First, scanning a map and creating a raster image filerequires no scientific or technical understanding of thegeologic product. Second, enhancing the raster image ofthe scanned map with computer software and creating araster image with standardized lines, text and graphic sym-bols is analogous to scribing and creating peel coats.Again, no scientific knowledge of geology is required toproduce the raster map image from an original map,

Digital Map Production and Publication byGeological Survey Organizations;

A Proposal for Authorship and Citation Guidelines

By C. R. Berquist, Jr.

Virginia Division of Mineral Resourcesc/o Department of Geology

College of William and MaryPO Box 8795

Williamsburg, VA 23187Telephone: (757) 221-2448

Fax: 757) 221-2093e-mail: [email protected]

although the replication process requires skilled labor. Theresult of the first two processes is only an image whetherraster or printed.

In the third procedure, GIS conversion requires sever-al steps including invention of a data structure, design andentry of multiple attributes, attention to spatial accuracy,and mathematical conversion of spatial data from one mapprojection and scale to others. Location error for all pointsin the GIS files is established by initial georeferencing thescanned map and by computer entry of each point, typical-ly achieved by the digitizer. For GIS conversion of previ-ously published maps, there will always be some horizon-tal error in the vector data because it becomes separatedfrom its original “married” topographic base. We havefound geologic features are most accurately located rela-tive to the original base only. Accurate entry of attributedata into the GIS files requires an interpretation of geolog-ic map data. There are also digital cartographic decisionsin the final production of a map image from the GIS (vec-tor) data. The GIS software is used to create tables ofalphanumeric spatial data and a final raster map image.Digital conversion of maps into and by a GIS program cre-ates the potential for errors in the GIS table data. Theseerrors in the GIS data files are not obvious by visualexamination of the image created from those GIS datafiles. The errors in the GIS data files are more difficult toexpose and repair than errors created through scanning anduse of computer graphics software. The result of the thirdprocess, (GIS conversion) is an image, raster or printed,and a number of files of alphanumeric GIS data.

Perhaps the relationship of the GIS files to the desiredend products obscures authorship issues. Some organiza-tions are interested in creating spatial data files from anexisting map and the files are the only end products.Other organizations create GIS files and then use the filesto create a final map (raster image and print). If the filesare published, both the files and the map are end products;if the files are not published, they are an intermediateproduct. Furthermore, the GIS files might be compared tothe scribed films and stick-up acetates (intermediate prod-ucts) of the traditional map-making procedure. If the filmsand acetates could have been “published”, who wouldhave been their author? Responsibility for collection of thegeologic data, synthesis of the data, creation of spatialdata, and assembly of the map should be attributed accu-rately and consistently for all map products, regardless ofthe format. This requires clearly stating who is responsi-ble for what.

PROPOSED RESOLUTION

During the search for an established authorship andcitation convention and in discussions with the USGS,librarians, and other state geologic surveys, a few groundrules became clear:

1. One who replicates (copies) original work withsoftware and creates only a graphic image does not requirethe same degree of experience or understanding of geolog-ic science as the original author. Replication contrasts tocompilation. Compilation is the creation of a new, modi-fied (from others), different work (map) requiring intellec-tual effort, experience and understanding of geology. Weshould maintain this traditional definition and use of“Compilation”. There is some confusion created becauseothers have used “Compilation” in new titles or authorshipof “digitally converted” older material and may have notmade the distinctions here described.

2. The digital files (spatial data) created from anysource map are different from the original map material(the image). The author of the digital files could havemade errors or omissions. Responsibility and culpabilityfor accuracy of the digital files cannot and should not beassigned solely to the original source author unless thesource author created the digital files.

3. Reference to the original geologic map and authorshould remain intact in any subsequent reworking or adap-tation.

4. Realize that every created work has a primaryauthor. Referencing and citing printed material follows anestablished convention. Referencing and citing digitalfiles and images has no established convention.

SUGGESTED CONVENTIONS

1. If a paper map is produced by a computer-graphicssystem which does not create digital data capable of beingqueried or searched, no advancement from the standardway of making maps has been gained and the process isanalogous to scribing. Cartographic credit may be givenas text printed on the image, but the originator (field geol-ogist, compiler) should retain authorship for the CD-ROM(raster image) and paper print. The term “Digital GeologicMap” should not be used.

2. If the map merely is scanned and an image formator formats made, retain the authorship of the originalsource on the published electronic media. The title of thework should be “Scanned geologic map of the….”, explic-itly excluding the word “Digital”; the original mylar orcompilation map could be assigned a “Manuscript MapXX” number or, if published, the formal map series num-ber. In the example below, John Doe was the geologistwho mapped Walkers quadrangle:

Doe, John, 1997, Scan of the geologic map of the Walkersquadrangle, Virginia: Virginia Division of MineralResources Manuscript Map 97-3 (unedited), [CD-ROM; 1997, July 5].

3. When GIS software is used to produce a mapimage and that image is published (raster or paper), prima-


ry authorship remains with the source author (originator).The title of a GIS-created paper map should read “DigitalGeologic Map of…”. Use of the word “digital” in thetitle clearly indicates GIS data was used and available.Furthermore, a clear distinction should be made betweenthe new GIS-created image and a previously publishedmap. Under the title on the digital map (image), list“Geology by John Doe” which gives full credit to theoriginal author for the original work and the map image,printed or raster. Under the author and in smaller font, listsomething like “Digital Conversion (and/or digital edit-ing) by Jane Smith” where Jane was the creator of thedigital files (GIS files). Because the map is not merely ascan of the original, the relator term “adapted from” isused in the citation:

Doe, J., 1998, Digital geologic map of the Walkers quad-rangle, Virginia: Virginia Division of MineralResources Digital Publication DP-5-A [CD-ROM;1998, June 21]. Adapted from John Doe, 1997, Scanof the geologic map of the Walkers quadrangle,Virginia: Virginia Division of Mineral ResourcesManuscript Map 97-3 (unedited), [CD-ROM; 1997,July 5].

If the map had not previously been open-filed or con-ventionally printed, the citation (Rader and Gathright arethe geologic authors) might simply read:

Rader, E.K., and Gathright, T.M., II, 1998, Digital geolog-ic map of the Front Royal 30 x 60 minute quadrangle,Virginia: Virginia Division of Mineral Resources Mapon Demand MOD-12 [1998, September 11].

4. The raster images and paper geologic maps createdfrom GIS spatial data require a different intellectual contri-bution than that required by computer graphics alone. Ifthe GIS files (tabular data) are published, cite this dataseparately from the map image by assigning a distinctivetitle (Digital Publication DP-5-B) and appropriate author-ship, but link the GIS data to the map image (DigitalPublication DP-5-A). In addition, relate the GIS data tothe source map by using “Adapted from…” in citations.As in the following example, the creator of the vector datais Jane Smith:

Smith, Jane, 1998, Geologic spatial data of the Walkersquadrangle, Virginia: Virginia Division of MineralResources Digital Publication DP-5-B [CD-ROM;1998, June 21]. Adapted from John Doe, 1997, Scanof the geologic map of the Walkers quadrangle,Virginia: Virginia Division of Mineral ResourcesManuscript Map 97-3.

5. When the original map author is also the creator ofthe GIS files and the original map has been published, thecitation could appear as in #6, below, or as:

Whitlock, W., 1998, Digital geologic map of the Gate Cityquadrangle, Virginia: Virginia Division of MineralResources Digital Publication DP-10-A [CD-ROM;1998, August 10]. Adapted from W. Whitlock, 1997,Geologic map of the Gate City quadrangle, Virginia:Virginia Division of Mineral Resources Open-FileMap 97-12.

There would be a companion publication for the GISvector data, cited as follows:

Whitlock, W., 1998, Geologic spatial data of the Gate Cityquadrangle, Virginia: Virginia Division of MineralResources Digital Publication DP-10-B [CD-ROM;1998, August 10]. Adapted from W. Whitlock, 1997,Geologic map of the Gate City quadrangle, Virginia:Virginia Division of Mineral Resources Open-FileMap 97-12.

6. When the mapping geologists compile their ownfield data on a computer with the GIS software (using theDRG topographic maps for a base), there is no mylar orintermediate product to open-file, unless it is the fieldnotes. Several of our staff in Virginia are currently creat-ing a final map and GIS vector files directly from fielddata. There could be one or two publications (titles) bythe mapping geologist(s):

Berquist, C.R., Jr., 1999, Digital geologic map and geolog-ic spatial data of the Williamsburg 30 X 60-Minutequadrangle, Virginia: Virginia Division of MineralResources Digital Publication DP-25 [CD-ROM;1999, November 21].

7. Shared authorship and the order of authors wouldfollow normal standards of mutual agreement betweenthose involved with creation of the particular work. Thisincludes consideration of the level of effort of digital com-pilers and digital editors for authorship of the digital files.Deceased geologic authors will gain a posthumous publi-cation for their geologic map (image) if this convention isadopted.

8. This proposed citation style is somewhat new, andusers of map products should be fully informed of thisconvention for authorship and citation. For example, toinsure better understanding, put the following text on theraster (and subsequently printed) map image as well as onthe CD-ROM jewel case:

DIGITAL MAP PRODUCTION AND PUBLICATION BY GEOLOGICAL SURVEY ORGANIZATIONS 41


Digital Publication 7 - Appalachia QuadrangleDigital Publication DP-7-A is the digital map (raster) image.The suggested bibliographic citation for this image of the geo-logic map or a print of this image is as follows:

Nolde, J. E., Henderson, Jr., and Miller, R. L., 1998,Digital geologic map of the Virginia portion of theAppalachia quadrangle: Virginia Division of MineralResources Digital Publication DP-7-A [CD-ROM;1998, August 26]. Adapted from Nolde, J. E.,Henderson, Jr., and Miller, R. L., 1988, Geology ofthe Virginia portion of the Appalachia and Benhamquadrangles: Virginia Division of Mineral ResourcesPublication 72.

Digital Publication 7B is the digital dataset. Geologic infor-mation, concepts, and other products gained from the use ofits files should be credited as follows:

Uschner, N. E., Jones, K. B., Sheres, D. E., and Giorgis, S.D., 1998, Geologic spatial data of the Virginia portionof the Appalachia quadrangle: Virginia Division ofMineral Resources Digital Publication DP-7-B [CD-ROM; 1998, August 26]. Adapted from Nolde, J. E.,Henderson, Jr., and Miller, R. L., 1988, Geology ofthe Virginia portion of the Appalachia and Benhamquadrangles: Virginia Division of Mineral ResourcesPublication 72.

9. Another situation to consider is when a digital mapis created from a mylar and this map image has never beenpublished; when the image is printed, it is the first andonly paper map available. This paper map can be printedand sold as a map-on-demand (MOD-n). The MOD seriesshould be restricted to products that will never have theassociated digital files published. Otherwise, there will beredundancy in the publication (series) nomenclature. Forexample, the same printed map could have a MOD-x num-ber and a DP-y number (x and y could be different or thesame, the result is that there are two series descriptors forthe same image). In the example below, the digital filesare unavailable (the map could have been produced bygraphics-based software). Put the following on the rastermap image and consequently the paper map:

The suggested bibliographic citation for this map is as fol-lows:

Rader, E. K and Gathright, T. M., II, 1998, Digital geolog-ic map of the Front Royal 30 x 60 minute quadrangle,Virginia: Virginia Division of Mineral Resources Map

On Demand MOD-9 [1998, September 11]. Adaptedfrom Rader, E. K. and Gathright, T. M., II, 1998,Scan of the geologic map of the Front Royal 30 x 60minute quadrangle, Virginia: Virginia Division ofMineral Resources Open-File Map 98-1 (unedited).

10. Revisions to publication series such as MODs andDPs are anticipated. When a revision is made to a prod-uct, add “revised <date> “ to the date in the citation. Asecond and additional revisions should be added withoutremoving the earlier dates, the same as done for books.

Berquist, C.R., Jr., Uschner, N.E., and Ambroziak, R. A.,2000, Spatial data of the digital geologic map ofVirginia: Virginia Division of Mineral ResourcesDigital Publication DP-14-B [CD-ROM; 1999, May5; revised 2000, July 5]. Adapted from VirginiaDivision of Mineral Resources, 1993, Geologic Mapof Virginia: Virginia Division of Mineral Resources,scale 1:500,000.

In summary, these suggestions rely on accepting thenotion that a geologic map image is different from GISspatial data, and that each work may receive a separatetitle. Relating new and original work by using a phrase“adapted from…” or “modified from…” will help insurecomplete and accurate citations. Reference to originalwork may require geological surveys to assign preliminarymaterials (drafted geology on topographic mylars, manu-script paper maps) to a series such as “Open-File Map” or“Manuscript Map”.

ACKNOWLEDGMENTS

I would like to thank Ian Duncan, William Whitlock,Eugene Rader, and Elizabeth Campbell from VDMR stafffor their review, comments, lively discussions and sugges-tions. Kelvin Ramsey of the Delaware Geological Survey,Elizabeth Koozmin and David Soller of the USGS, JamesCraig of Virginia Tech, Kathryn Blue, Patricia Hauseman,and Karen Berquist of Swem Library, College of Williamand Mary also provided helpful criticism and suggestions.

Three online sources of information were also helpful:<http://www.swem.wm.edu/Gateway/citations.html><http://www.uvm.edu/~ncrane/estiles/apa.html><http//www.sciam.com/0796issue/0796okerson.html>.

43

Mention metadata to people and you’ll likely see themcringe. They cringe because metadata lies at the interfacebetween people who create spatial data and the peoplewho would use those data, and it is hard to communicateto a prospective data user all of the wisdom needed tomake proper and effective use of the data. Metadatafocuses our attention on difficult educational, cultural, andtechnological issues inexorably linked with the commondesire of scientists to see their work valued by other peo-ple.

Figure 1 shows how metadata are transferred fromdata producers to data users. In small, active researchgroups it is possible to communicate using mostly techni-cal terminology, both because the data producers and datausers share a common understanding of the terminologyand the circumstances under which the data were producedand because the data users are able to return to the dataproducers frequently to ask questions. The situationchanges drastically when data are made available throughthe internet to people who are not well known by the dataproducers. Data users may have different backgroundsand needs, and there are many sources of data available.Consequently data users need metadata that are readilysearchable and can be presented in a variety of differentformats.

Metadata are readily searchable only if they conformto a standardized structure; they can be presented in a vari-ety of different formats only if they are parseable by com-puter software. These needs require the metadata to bedescribed using technical terminology that is standardizedacross many scientific and technical disciplines. But thattechnical terminology is often unfamiliar to specialists inany one discipline. Hence metadata are not only hard forproducers to write, they are also hard for reviewers andend-users to read. Plain-language approaches providesolutions directed at easing specific troubles encounteredin writing and reading metadata.

PLAIN-LANGUAGE APPROACHES TOCREATING, REVIEWING, AND READINGMETADATA

1. Understanding Metadata and LearningHow to Write it

Metadata in Plain Language: A guide for metadatacreators and reviewers:

Plain-Language Resources forMetadata Creators and Reviewers

By Peter N. Schweitzer


Telephone: (703) 648-6533Fax: (703) 648-6560


Figure 1. Diagram depicting the flow of metadata fromdata producers to data users. This information transfercannot occur using plain language alone because unstruc-tured information cannot be reliably indexed by topic,place, time, and other characteristics, nor can it be reliablyreexpressed in a variety of formats suiting the needs ofdiverse users. Consequently the metadata must beexpressed using some technical terminology in a standard-ized structure and parseable format; these allow computersoftware to index it appropriately and reexpress the meta-data in plain language as well as technical terminology.

<http://geology.usgs.gov/tools/metadata/tools/doc/ctc/>

The goals of these pages are to put the metadata stan-dard’s many elements into perspective and to show how tocreate a metadata record using a heuristic procedure. Theypresent the FGDC metadata standard as a series of plain-language questions; for each question they show whichelements of the metadata standard should hold theanswers, and how to express the answers correctly.

2. Expressing Metadata as Answers toStandard Questions

The USGS metadata parser mp reads, checks, and re-expresses metadata:

<http://geology.usgs.gov/tools/metadata/tools/doc/mp.html>

MP checks the structure of metadata against theFGDC standard, indicating where and how the metadatarecord doesn’t conform. MP then re-expresses the metada-ta in a variety of formats that are likely to be useful in dif-ferent ways. MP can create SGML, XML, parseable text(its input formats), as well as HTML and DIF, a form usedby the NASA Global Change Master Directory. Of partic-ular interest is the new FAQ-style HTML, which presentsthe metadata as answers to the same plain-language ques-tions given in Metadata in Plain Language. The abbrevia-tion FAQ as commonly found on the internet refers to fre-quently-asked questions; since metadata are producedbefore a data set is used extensively by the public, FAQhere refers to frequently-anticipated questions. This formis likely to be more readable by people who are unfamiliarwith the FGDC standard.

How to generate FAQ-style HTML using mp:

The simplest method is to invoke mp with the -foption:

mp info.met -f info.html

This will place into the file info.html the FAQ-style HTML output.

An alternative method is to specify the form of the filename in a config file. Under output:html the name of

the FAQ-style HTML output is specified using the elementfaq as follows:

output:html:

file: %s.html

faq: %s.faq.htmlHere the file element contains a template showing

how mp should compose the name of the outline-styleHTML file; in the template, the string %s is replaced withthe name of the input file (with its extension removed).Likewise, the faq element contains a template showinghow mp should compose the name of the FAQ-styleHTML file.

Note that the Clearinghouse server software currentlyin use (May-1999) assumes that if the SGML metadatadocument selected is info.sgml (for example) then theHTML document to be returned to the user is info.html.That strategy will normally return the outline form ofHTML to the user. If you wish to return the FAQ-styleHTML first, then you should change the values shownabove like this:

output:html:

file: %s.out.htmlfaq: %s.html

Using these config file elements, the FAQ-style outputwill be the one returned by the server through a Z39.50PRESENT request.

NOTE: mp now provides in its HTML output a linkto each of the other output formats that you requestedwhen running mp. These links are relative to the currentdirectory by default, and will work correctly when some-one retrieves a metadata record directly through a webserver. However, HTML metadata records retrievedthrough the Clearinghouse gateway interface come taggedwith the URL of the gateway, consequently these links willnot work by default with HTML records found through thegateway interface. To make these links work withoutregard to the retrieval method, place a BASE tag into theHEAD element of the output HTML code. As you mightguess, mp can do this for you, but it needs to know theURL where your metadata will be available as web pages.It gets this information from a config file entry as follows:

output:html:

base: URL

So if your web site has a URL like

<http://www.our-data.org/metadata/>

that will contain your metadata records, put this into yourconfig file:

output:html:

base: http://www.our-data.org/ metadata/


Obviously you have to use the -c config_file com-mand line option for mp, substituting for config_file thename of the actual config file you’ll be using.

How mp writes FGDC metadata elements in plainlanguage:

<http://geology.usgs.gov/tools/metadata/tools/doc/plain.faq.html>

This is a specially constructed record showing howmp composes plain-language output from FGDC elements.Generated entirely by mp, it shows where the elements ofthe metadata standard appear in the output as answers toquestions. This file should not be regarded as authoritative,since there are many choices that must be made when cre-ating metadata, however it may help people to understandhow mp composes answers to the plain-language questionsusing the standard FGDC elements.

3. Examples Showing Plain-Language andTechnical Terminology

Geology of the onshore part of San Mateo County,California: A digital database

Questions & Answers (FAQ-style HTML)<http://geo-nsdi.er.usgs.gov/metadata/ofr98137.faq.html>

Outline (original-recipe HTML)

<http://geo-nsdi.er.usgs.gov/

metadata/ofr98137.html>Parseable text (not HTML)

<http://geo-nsdi.er.usgs.gov/

metadata/ofr98137.met>

Additional examples can be found on the U.S.Geological Survey Geoscience Data node of the NationalGeospatial Data Clearinghouse, at <http://geo-nsdi.er.usgs.gov/>.

PLAIN-LANGUAGE RESOURCES FOR METADATA CREATORS AND REVIEWERS 45

ArcInfo Version 8 contains the same core applicationsas Version 7.x. These applications—ARC, ARCPLOT,and ARCEDIT— have been enhanced and updated. Thisensures binary compatibility with any application devel-oped using Version 7.x. Three new applications, ArcToolbox, Arc Catalog, and Arc Map, provide easy-to-use,Windows-style user interfaces that make working withArcInfo easier than ever before. The new componentobject data model introduced in Version 8 goes beyond tra-ditional point, line, and polygon features and allows mod-eling of objects, such as power poles and electrical switch-es, in a more realistic way. These new applications arehighly customizable.

Information about the Environmental SystemsResearch Institute, Inc. (ESRI) and its products, includingArcInfo and ArcView, can be found at<http://www.esri.com>. A short summary of new func-tionality in ArcInfo Version 8, as published in ArcUsermagazine, is available at<http://www.esri.com/news/arcuser/0499/ai8b.html>. Forinformation on application of ESRI products in the earthsciences (e.g., collaboration with USGS on development ofsymbolsets and patternsets for geologic features) or tooffer suggestions or discussion, please contact me.

47

ArcInfo 8: New GIS Technology from ESRI

By Mike Price

Environmental Systems Research Institute, Inc.380 New York AvenueRedlands, CA 92373

Telephone: (909) 793-2853 ext. 1677Fax: (909) 307-3037


49

As commercial companies such as Dynamic Graphics,Inc. (developer of the 3D geologic sturcture and property-modeling system, EarthVisionTM), universities, the variousgeological surveys, and other research organizations haveextended traditional mapping techniques toward creationof true 3D geologic structure and property models, accessto the information in those models for external uses hasnot been as good as it ultimately needs to be. One of theprimary benefits of a 3D geologic structure modeling sys-tem is the enforced consistency of the component zones,fault blocks, and surfaces that comprise the model. Such amodel is easily viewed on a computer, picked apart, sliced,

rotated and so on . . ., but creating maps, sections, andother electronic outputs from such a model is very impor-tant if that model is to be more than a nice curiosity toanyone other than those who can view it on a computer.In this talk, I will review the process of creating suchmaps, sections and outputs, and discuss why such mapsrepresent an improvement over maps developed indepen-dent of an underlying model.

For further discussion on 3D models and their rela-tionship to mapping outputs, please refer to<http://www.dgi.com> and/or contact Skip Pack (email:[email protected]).

Extracting the Information from 3D Geologic Models —Maps and Other Useful Outputs

By Skip Pack

Dynamic Graphics Inc.1015 Atlantic Avenue

Alameda, CA 94501-1154Telephone: (510) 522-0700


51

The Ground Water Atlas of the United States presentsa comprehensive summary of the Nation’s ground-waterresources, and is a basic reference for the location, geogra-phy, geology, and hydrologic characteristics of the majoraquifers in the Nation. The information was collected bythe U.S. Geological Survey (USGS) and other agenciesduring the course of many years of study. Results of theU.S. Geological Survey’s Regional Aquifer-SystemAnalysis Program, a systematic study of the Nation’smajor aquifers, were used as a major, but not exclusive,

source of information for compilation of the Atlas. TheAtlas includes 14 chapters, 13 of which are published sep-arately, representing regional areas that collectively coverthe 50 States and Puerto Rico. These chapters are pub-lished in the Hydrologic Atlas series with numbers rangingfrom HA-730B through HA-730N. The remaining chap-ter, Chapter A, serves as an introduction and national sum-mary. Chapters will also be collated into a single boundvolume (figure 1).

Ground Water Atlas of the United States

By Gary D. Latzke

U.S. Geological Survey505 Science Drive

Madison, WI 53711-1061Telephone: (608) 238-9333 x141


Figure 1. Organization of the Ground Water Atlas of the United States.

ATLAS ORGANIZATIONThe Ground Water Atlas of the United States is divided into 14 chapters. Chapter A pre-sents introductory material and nationwide summaries; chapters B through M describe allprincipal aquifers in a multistate segment of the conterminous United States; and chapter Ndescribes all principal aquifers in Alaska, Hawaii, and Puerto Rico.

Chapter content HydrologicAtlas Chapter

– Introductory material and nationwide summaries 730-A1 California, Nevada 730-B2 Arizona, Colorado, New Mexico, Utah 730-C3 Kansas, Missouri, Nebraska 730-D4 Oklahoma, Texas 730-E5 Arkasas, Louisiana, Mississippi 730-F6 Alabama, Florida, Georgia, South Carolina 730-G7 Idaho, Oregon, Washington 730-H8 Montana, North Dakota, South Dakota, Wyoming 730-I9 Iowa, Michigan, Minnesota, Wisconsin 730-J10 Illinois, Indiana, Kentucky, Ohio, Tennessee 730-K11 Delaware, Maryland, New Jersey, North Carolina 730-L

Pennsylvania, Virginia, West Virginia12 Connecticut, Maine, Massachusetts, New Hampshire 730-M

New York, Rhode Island, Vermont13 Alaska, Hawaii, Puerto Rico 730-N


The main objective of the Ground Water Atlas was toproduce an attractive and easy to follow product with text,charts, photos, cross sections and diagrams which explaintechnical terms in simple language. It attempts to clearlyillustrate the principles that govern the movement andoccurrence of ground water in different geologic and topo-graphic settings. To accomplish this task, hydrologistsfrom the Water Resource Division of the USGS wereassigned to each of the 14 chapters based on their knowl-edge of the ground water in the area. The authors com-piled from previously published reports. Graphics andmaps from a variety of sources were used or modified andoriginal graphics were added where needed. Graphics andtext were forwarded to a USGS office, the Cartographyand Publishing Program (CAPP) in Madison, Wisconsin,for publication production.

The coordination of design for the atlas series was anenormous task, selecting colors, patterns and cartographicsymbolization for each aquifer and insuring that thosedesign decisions made in the early chapters could beapplied to all 14 chapters. In its history, production meth-ods for the Ground Water Atlas have changed from exclu-sively traditional to exclusively digital cartographic meth-ods. A publication requirement was to make all chaptersappear like a single product regardless of the productionmethods used. The first two chapters, 730-G and 730-J,were done using traditional cartographic methods, needingscribe coats, peel coats, type flaps and many hours ofdarkroom work to produce. The following two chapters,730-H and 730-C, had incorporated digital elements intheir production. Basemap line work was created inArc/Info using USGS DLGs imported into AdobeIllustrator on the Macintosh platform for annotation.Other simple graphics were completed in Illustrator aswell and imaged to a page-size image setter. With theaddition of a large size film plotter in 1993, the next chap-ter, 730-K, and all subsequent chapters have been pro-duced digitally. Arc/Info was used for all base and aquiferoutcrop data, Adobe Illustrator for figure production,Adobe Photoshop for photo manipulation, and AdobePageMaker for text, layout and creating separates.Producing these chapters digitally has saved time in pro-duction, eliminated almost all darkroom work and cut thetime spent making corrections by 80 percent.

As of mid 1999, 12 of the 13 regional chapters havebeen published and are available. Chapter 730-N is

expected to be published late in 1999 (figure 2). Thebound hard cover volume of all 14 chapters is also expect-ed to be released in fiscal year 2000. Several chapters areavailable at the following web site:<http://wwwcapp.er.usgs.gov/publicdocs/gwa/index.html>.This web site, which is currently under construction, willcontain all 14 chapters when completed. Chapters pro-duced using traditional cartographic methods will be avail-able for viewing as JPEGs and downloadable in TIFF fileformat. Chapters produced using digital cartographicmethods will be viewable as GIF files and downloadableas compressed EPS files.

As a result of the Ground Water Atlas activities, anational map, the “Principal Aquifers of the UnitedStates”, and an associated data set have been publishedand are available as part of the “National Atlas of UnitedStates ™”. The data and map describe the distribution ofprincipal aquifers across the country with each aquiferbeing classified as one of six geologic material types. Thedata and printed map are available from the “NationalAtlas of United States” web site at: <http://www-atlas.usgs.gov/atlasmap.html>. The aquifer data is down-loadable in several formats including SDTS, shape files orARC export file format. Links to the aquifer data andmetadata can be found from either the Ground Water Atlasor National Atlas of the US web sites.

Figure 2. Status of Ground Water Atlas chap-ter publication.

Availability of the Ground Water Atlas ofthe United StatesHydrologic Atlas Publicationchapter status, May, 1998

HA 730-B published 1995HA-730-C published 1995HA-730-D published 1997HA-730-E published 1996HA-730--F published 1998HA-730-G published 1990HA-730-H published 1994HA-730-I published 1996HA-730-J published 1992HA-730-K published 1995HA-730-L published 1997HA-730-M published 1995HA-730-N published 1999

INTRODUCTION

The Kentucky Geological Survey (KGS) is currentlyin the process of converting a statewide set of geologicquadrangle (GQ) maps to digital format (Anderson, 1998).Since geologic maps are usually highly detailed and com-plicated, the KGS devised a standardized procedure forGQ data capture and attribution, in which data for a partic-ular quadrangle are segregated into multiple Arc/Info datalayers based on the nature of the geologic data (Andersonand others, this volume). For each GQ, lithologic forma-tions, structure contours, mineral resources, faults, coalresources, fossils, drill holes, and igneous dikes are cap-tured and attributed as separate coverages or data layers.Each data layer may consist of one or more feature datatypes (e.g., polygon, arc, or point). These coverages arestored under an Arc/Info workspace named after the quad-rangle. There are 707 geologic quadrangle maps thatcover Kentucky. Eventually, the KGS digital GQ databasewill contain 707 Arc/Info workspaces and several thousandArc/Info coverages. Therefore, populating data from thisspatial database for map manipulation is an importantissue. So that users of the digital GQ database will nothave to browse through the database to find particular cov-erages, and then load them one by one, we introduce anew ArcView application, called GQ Data Query, thatacquires spatial geologic data from a spatial databasethrough an interactive index map.

ALGORITHM OF GQ DATA QUERY

GQ Data Query can significantly automate access to,and display of, digital GQ data. It allows a user to loadgeologic data for any quadrangle by selecting that quad-rangle in a digital index map. The algorithm of GQ DataQuery is analogous to acquiring a paper GQ map from an

organized map storage. The algorithm can be illustrated asthe following basic steps: (1) locate a quadrangle byselecting the polygon that represents the quadrangle on theindex map, (2) retrieve the quadrangle name by accessingthe attribute table of the index coverage, (3) identify thedata source where the Arc/Info workspace for the selectedquadrangle is located, (4) identify all existing Arc/Infocoverages under the workspace, and (5) extract and displayall the Arc/Info coverages under the workspace (Fig. 1).

53

Acquisition of Geologic Quadrangle Data from a Digital IndexMap—A New ArcView Application to Geologic Data Query

By Xin-Yue Yang, Warren H. Anderson, Thomas N. Sparks, and Daniel I. Carey

Kentucky Geological SurveyRoom 228, Mining and Minerals Resources Building

Lexington, KY 40506Telephone: (606) 257-5500


Figure 1. Algorithm and implementation of GQData Query.

Developed through the Avenue scripting language, GQData Query consists of the following major components: a7.5-minute index map coverage, an external database filethat stores the metadata for all individual quadrangles, anda set of Avenue scripts. The Avenue scripts are customizedto associate with a button and the Load Data menu underView in the standard ArcView 3.1 interface, so that a userwill be able to load all geologic data pertinent to a selectedquadrangle by either pushing the button or selecting themenu. The database file is used to link the polygonattribute file of the index polygon coverage to form a vir-tual table, which then serves as a linkage between theindex coverage and the actual GQ data. A user can alsoobtain metadata for any particular quadrangle through thevirtual table, such as quadrangle name and number, lati-tude and longitude, publication date, and author name. Ifthe index map is overlaid with other coverages such ascounty boundaries and road networks, a user will be ableto identify the spatial relationship between a quadrangle,road network or county area.

EXAMPLE

Figures 2 and 3 illustrate a typical working session ofGQ Data Query. When the ArcView project is opened, a7.5-minute index map is displayed. On the index map,some quadrangles are highlighted initially to indicate thatdigital geologic data for these quadrangles are currentlyavailable (Fig. 2, top). The desired quadrangle can beselected either through query by using the Query Builder(by name or map number) or through mouse click afterzooming in (Fig. 2, bottom). After a quadrangle is select-ed on the index map, the geologic data can be loaded sim-ply by pushing the Load GQ Data button or selecting LoadData menu (Fig. 3, top). If users want to extract data formultiple quadrangles, they will be given an option to pop-ulate data on an existing view if quadrangles are adjacentto each other (Fig. 3, bottom) or create a new view.

REFERENCEAnderson, W.H. 1998, History of geologic mapping at the

Kentucky Geological Survey, in Soller, D.R., ed., DigitalMapping Techniques ‘98-Workshop Proceedings: U.S.Geological Survey Open-file Report 98-487, p. 9-12,<http://pubs.usgs.gov/openfile/of98-487/anderson.html>.


ACQUISITION OF GEOLOGIC QUADRANGLE DATA FROM A DIGITAL INDEX MAP 55

Figure 2. A typical working session of GQ Data Query: (top) an interactive index map as a graphic inter-face of ArcView GIS; (bottom) select a quadrangle on the index view.


Figure 3. A typical working session of GQ Data Query (continued): (top) loading a geologic quadrangle;(bottom) loading two adjacent quadrangles.

In August 1996, representatives of the Association ofAmerican State Geologists (AASG), the GeologicalSurvey of Canada (GSC), and the U.S. Geological Survey(USGS) met in St. Louis, MO, to discuss the developmentof various standards and guidelines for geologic maps con-tributed to the National Geologic Map Database(NGMDB). As a result, working groups were formed toaddress geologic map symbology, data capture, metadata,map publication, and a conceptual data model (see<http://ncgmp.usgs.gov/ngmdbproject/>). The Data ModelWorking Group began its work shortly thereafter and,through numerous public presentations, discussions, andprogress reports (for example, Raines and others, 1997;Soller and others, 1998), in October 1997, the WorkingGroup posted to the NGMDB project’s Web site for publiccomment a report describing the proposed standard datamodel. The initial public version was superceded by amore mature one in May 1998 (version 4.2, Johnson andothers, 1998a). A technical review of model version 4.2was conducted at a 3-day workshop in June 1998. Minorrevisions subsequently were made, and the most currentversion, 4.3 (Johnson and others, 1998b), was released.

At the Digital Mapping Techniques ‘98 workshop, thedata model was discussed extensively, and various agen-cies expressed interest both in furthering its conceptualdevelopment and in testing its implementation with their

geologic maps. We are pleased to note that many of thepapers given at this meeting (DMT’99) provide evidenceof this committment, and of their interest in continuing theeffort to develop a standard data model for the geosciencecommunity.

With technical review and release of version 4.3, theData Model Working Group had completed its task. Theagencies involved in this effort then devised a mechanismto promote further development of the conceptual modeland the various implementations that would be requiredamong the participating agencies. This mechanism, theData Model Steering Committee, provides overall guid-ance, coordination, publicity, and communication for thedevelopment of a digital geologic map data model to sup-port, at a minimum, the needs of the United States andCanadian geoscience community. In early 1999, the firstmeeting was held, and a charter was written (see<http://geology.usgs.gov/dm/steering/charter.html>). Thesecond meeting was held at the DMT’99 workshop, andincluded the development of plans for various technicalteams to conduct specific tasks. These teams will address:

- Requirements Analysis (to refine our understanding ofthe data analysis requirements of various users);

- Data Model Design (to continue refining the conceptualmodel based on the Requirements Analysis, delibera-

57

Progress Toward Development of a Standard Geologic MapData Model

By the Geologic Map Data Model Steering Committee

Members:Brian Berdusco (Ontario Geological Survey, [email protected])

Tom Berg (Ohio Geological Survey, [email protected])Boyan Brodaric (Geological Survey of Canada, [email protected])

Jim Cobb (Kentucky Geological Survey, [email protected])Bruce Johnson (U.S. Geological Survey, [email protected])

Rob Krumm (Illinois State Geological Survey, [email protected])Jon Matti (U.S. Geological Survey, [email protected])

Scott McColloch (West Virginia Geological and Economic Survey, [email protected])Gary Raines (U.S. Geological Survey, [email protected])

Peter Schweitzer (U.S. Geological Survey, [email protected])Dave Soller (U.S. Geological Survey, [email protected])

Loudon Stanford (Idaho Geological Survey, [email protected])

tions of the other technical teams, and user com-ments);

- Scientific Language Technical Team (to develop standardterminologies for the various elements that comprisegeologic maps, e.g., rock classification);

- Software Tool Development (to design tools that meetuser needs as specified in the Requirements Analysis);

- Data Interchange (to develop translators among variousimplementations of the conceptual model);

- Documentation (to improve public understanding of datamodel design and software tools).

Information concerning the Steering Committee andthe technical teams are posted to the data model Web con-ference site, <http://geology.usgs.gov/dm/>. Interestedpersons are invited to register at the site and contribute tothe data model’s continued evolution.

REFERENCES

Johnson, B.R., Brodaric, Boyan, and Raines, G.L., 1998a, DigitalGeologic Maps Data Model, v. 4.2: AASG/USGS DataModel Working Group Report, <http://geology.usgs.gov/dm/>.

Johnson, B.R., Brodaric, B., Raines, G.L., Hastings, J., and Wahl,R., 1998b, Digital Geologic Map Data Model v. 4.3:AASG/USGS Data Model Working Group Report,<http://geology.usgs.gov/dm/>.

Raines, G.L., Brodaric, Boyan, and Johnson, B.R., 1997,Progress report—Digital geologic map data model, in DavidR. Soller, ed., Proceedings of a workshop on digital map-ping techniques: Methods for geologic map data capture,management, and publication: U.S. Geological SurveyOpen-File Report 97-269, p. 43-46,<http://ncgmp.usgs.gov/pubs/of97-269/raines.html>.

Soller, D.R., Brodaric, Boyan, Hastings, Jordan, Johnson, B.R.,Raines, G.L., and Wahl, R.R., 1998, Progress TowardDevelopment of a Standard Geologic Map Data Model, inDavid R. Soller, ed., Digital Mapping Techniques ‘98 —Workshop Proceedings: U.S. Geological Survey Open-FileReport 98-487, p. 47-48,<http://pubs.usgs.gov/openfile/of98-487/soller3.html>.


ABSTRACT

Digital geologic maps are a cornerstone of next gener-ation geologic information systems that will archive,query, retrieve, and display geologic information tailoredto specific requirements. These systems will probablyinclude a knowledge base component to support applica-tions that check the consistency of existing ‘knowledge’with new interpretations and data. In order to maximizethe utility of such a system, a standard conceptual datamodel must be developed. The purpose of a conceptualmodel is to provide a consistent framework for developinglogical and physical models specific to particular soft-ware/hardware systems. It is difficult or impossible totransfer data between information systems implementedusing inconsistent underlying concepts. Many subsets ofearth science overlap with other disciplines, and develop-ment of conceptual models in these domains can takeadvantage of similar efforts in other fields. Two domainsof discourse that are particular to earth science are 1) thedescription and classification of rocks (lithology), and 2)the description of spatial relationships necessary to deducethe relative ages of rock units and structures (map-scalerelationships). A conceptual model of these domainsdescribes the component parts, and the relationshipsbetween the parts.

Rock hand samples are small relative to our bodies (1-20 cm), and contain a very large number of constituentparts (grains). These parts can be classified into differenttypes characterized by their size, composition, and shape.Hand-sample lithology is described in terms of a set ofconstituent parts defined by generalized characteristics,and the typical relationships between the types of con-stituents. A complete lithology description defines the setof constituent part types, the fraction of each type in thewhole (modal petrography), and the typical relationships

between grains of each type. The set of constituent typesmay be defined based on one or more criteria involvingsize, shape and composition. The composition of a con-stituent type may be a single mineral or another lithologicentity (e.g. clasts in conglomerate, leucosome inmigmatite). Description of the relationships betweengrains includes information about the texture and fabric ofthe aggregate. Common lithologic terms may specifycharacteristics of only one aspect of the aggregate: e.g.grain size (sandstone), grain shape (breccia), fabric(schist), or grain composition (hornblendite).

On a ‘map scale’ (1-100 km), geologic objects arelarge compared to our bodies, and are internally variable.Earth features at map-scale are described in terms of a setof rock bodies defined by generalized characteristics, andthe surfaces that bound the rock bodies. A set of necessaryand sufficient conditions is required to define the identityof each rock body. Each rock body is bounded by a set ofintersecting surfaces that define a volume at a particulargeographic location. In many cases, one of the boundingsurfaces is the Earth’s surface. The other bounding sur-faces are referred to as geologic surfaces. These may bedepositional, intrusive, gradational, faulted, or polygenetic.Geologic surfaces are typically directed. A depositional orintrusive contact is directional if rock on one side isknown to be older than rock on the other side. A fault isdirected if the orientation of the slip vector is known. Agradational contact can have a sense defined by the gradi-ent of some physical property. Polygenetic contacts maynot have an inherent directionality, or they may have oneor more directions inherited from stages of their genesis.Geologic surfaces commonly have properties, for exampleboundary layer thickness, rock types unique to the bound-ary (fault rocks, soil profile, basal conglomerate...), andsmall-scale geometry of the surface (rough, smooth, inter-leaved, etc.).

59

Geologic Concept Modeling, with Examples forLithology and Some Other Basic Geoscience Features

By Stephen M. Richard

Arizona Geological Survey416 W. Congress, #100

Tucson, AZ 85701Telephone: (520) 770-3500


INTRODUCTION

Geologic data are used for land-management decision-making, engineering design, in the search for mineralresources, and for scientific research. Traditionally, geo-logic information has been stored and disseminated usinggeologic maps and written reports (Bernknopf et al.,1993). Because of the complexity of the earth, much ofthe information included in a geologic map is buried inseveral layers of abstraction. Production of derivativemaps designed for a specific purpose thus requires a geo-logically sophisticated analysis of the original map and thedrafting of a new map designed to depict a different aspectof the geology. Such maps might be designed to showrocks of a particular age, show the lithology of the rockswithout respect to age, show the orientation of bedding orfoliation in layered rocks, or to show the acid bufferingcapacity of the rocks, etc. Modern data storage and com-munication technology has created an opportunity torethink the manner in which geologic information isarchived and presented. This report discusses some ideasfor the conceptual schema of a computer-based geologicinformation system.

This system will need to meet the needs of land man-agers or planners needing information pertinent to regula-tory, planning, and development functions, mineral explo-ration geologists, researchers in search of detailed techni-cal information, and curiosity-driven users from the gener-al public. Many of these users may not be expert geolo-gists, but still need to be able to query the system to obtaininformation. The underlying data model must be flexibleenough to encompass a wide range of earth science infor-mation, storing it in such a fashion that it does not becomeobsolete with advances in geologic science.

As a starting point, it is useful to consider a fantasyview of the product. The ultimate geologic informationsystem would contain a detailed 3-dimensional model ofthe Earth. Such a description would include:

* quantitative descriptions of the chemical composition ofall materials

* quantitative descriptions of the physical characteristicsof materials at any scale

* a classification of the materials into lithostratigraphic,biostratigraphic, and chronostratigraphic rock units (ormaterial bodies).

* P-T-t paths for points throughout the model,

* descriptions of the static geometry and kinematic historyof structures

* dynamic descriptions of active tectonic processes

* the geologic history of all rock bodies

The ultimate interface to such a system would allownatural language queries couched in technical or nontech-nical terms, and respond by providing appropriate naturallanguage answers, data tables, standard format output files,or visualizations (maps, cross sections, 3-D views, etc.).Knowledge of the physical location and structure of thestored data would be unnecessary to the user. Multipleworking hypotheses would be stored for regions in whichknowledge is incomplete or inconsistent. The origin ofany particular fact or interpretation could be traced to itssource. The system will play the role presently filled bygeologic maps—providing information in response toqueries, checking consistency of data to be introduced intothe system, and archiving data for future use. The flexibil-ity of a computer-based system provides for a much morecomplete description of the Earth in a form that permitsmore sophisticated analysis, and can be user-customized tomeet particular needs.

Meanwhile, back in the real world of incompleteknowledge, limited budgets, and existing computer soft-ware/hardware environments, the geologists of today needto lay the groundwork for the data archive and analysissystem of the future. Standard engineering procedures forsystem design begin with a requirements analysis.Without going into detail, the desired geologic informationsystem must: 1) allow geologists to archive earth sciencedata and interpretations in a logically consistent form; 2)track data sources and updates to the data archive, and 3)provide applications that query, retrieve, and display exist-ing geologic information to meet the needs of environmen-tal, engineering, exploration, and research geoscientists.The full information system thus consists of a data reposi-tory and a set of applications (interfaces) specific to partic-ular fields of enquiry. The data repository subsystem con-sists of a data model to organize information, applications(methods) to track updates and maintain data integrity, andthe physical data storage implementation. This paper isconcerned with the design of the data repository datamodel for a geologic information system.

The design of the data model begins with develop-ment of a conceptual model that describes earth science(the universe of discourse) at a fundamental level usingearth science terms and concepts. The purpose of develop-ing a conceptual model is to provide a consistent frame-work for developing logical and physical models couchedin database-system terms and concepts. These guideimplementation of the information system in a particularcomputer environment. It is difficult or impossible totransfer data between information systems based on mod-els with inconsistent underlying concepts. For example, ifrocks were thought of only in terms of their whole-rockchemistry in one model, there would be no unique way toconvert data stored in that model into one in which rockswere modeled as aggregates of minerals.


The following sections analyze two key sub-domainsof earth science, geologic maps and lithology, and propos-es conceptual models for some aspects of these domains.The appendices include a glossary of terms used, a discus-sion of the philosophy of conceptual modeling, and areview of some existing data modeling efforts relevant tothe development of a geologic information system.

DOMAIN ANALYSIS:DESCRIPTION OF THE EARTH

The Earth is a composite object with a shape definedby solid matter (this discussion is limited to a staticmodel). It is composite because it consists of many defin-able parts. Its shape is a spheroid defined by the solidmatter of the lithosphere. The complexity of the Earth isqualitatively similar no matter what the scale of observa-tion—TEM, microprobe, thin section, hand sample, out-crop, quadrangle, province, continent, planet. For anyscale of observation, heterogeneity at smaller orders ofdimensional magnitude can be averaged, and heterogeneityat larger orders of dimensional magnitude is outside thedomain of interest. Thus, any model for describing theEarth must be hierarchical and recursive. A part of theEarth characterized by average properties at one scale canbe analyzed into component parts (with different averagedproperties) at a more detailed scale. The description isrecursive because some parts of a material might be mate-rials defined at the same level in the hierarchy. For exam-ple, the parts that make up a conglomerate (a rock type)are other rocks, maybe even fragments of a different con-glomerate.

In this discussion, the smallest scale of observationconsidered is the hand sample—that is pieces of rock theaverage person can hold in their hand and examine with amagnifying glass. Generally this means that aggregates ofparticles smaller than about 0.1 mm are represented byaverage properties, and particles larger than about 10 cmare entities. The next scale considered is the outcrop—theaverage size of exposed rock that can be examined anddescribed at one time, say 10 cm to 100 m. Map scale istaken to encompass 100 m to 100 km, and is the largestscale of observation considered here. The next larger ‘spa-tial domain’ might be called province scale, encompassingdescription of bodies with dimension 100-10,000 km.

Geologic descriptions can be associated with a pointof observation, or can be a generalized description appliedto a surface (fault, contact...) or rock volume (rock unit,region on a map, Formation...). For most of the commonlydescribed geologic features, there are systems of classifica-tion based on certain characteristic features. Rock and fos-sil names are examples of such classification systems. Acontact between rock bodies may be described as intrusive

or depositional, connoting certain features of the contact.Many observations made by geologists in the field aresummarized by identifying an observed object as a mem-ber of a predefined class. Such descriptive ‘data’ is sub-jective because assignment to a class depends on the clas-sification system chosen and the experience of the geolo-gist making the observations.

Part of the geologist’s work in mapping an area isdeveloping a system of classification for rocks, contacts,and structures that summarizes observations and simplifiesthe task of describing the variety of things observed.Some of the observed things will match criteria for stan-dard geologic classifications or classifications developedin nearby areas. When something can not be matched tosatisfaction with a known type, a new classification isdefined based on observations in the map area. The defin-ition must include a set of necessary and sufficient condi-tions to assign membership to the class, and place the classwithin the existing hierarchy of known things. Recordinga full description of the classification system used topigeon hole observations is an essential part of making ageologic map.

The Glossary of Geology (Bates and Jackson, 1987)defines ‘rock’ as an aggregate of one or more minerals, ora body of undifferentiated mineral matter (e.g. glass), or ofsolid organic matter (e.g. coal). This definition is modi-fied slightly here to define a rock as a consolidated aggre-gate of constituent parts, or a solid body of undifferentiat-ed mineral or solid organic matter. The constituent partsof a rock may be minerals or other rocks (as in a conglom-erate). A simple rock consists of an aggregate of distinctmineral grains (granite). More complex rocks may consistof aggregates of rock fragments (conglomerate, breccia),or mixtures of distinct lithologic components (migmatite,gneiss).

The system of hand-sample description proposed herefor use in computer databases is based strictly on featuresvisible in the hand sample that are descriptive and non-genetic. Bates and Jackson (1987) define lithology as thephysical character of a rock. In this report the term lithol-ogy will be modified by terms indicating the scale ofdescription, and will refer only to rock-volume characteris-tics. The hand-sample scale classification scheme forEarth materials makes a first order distinction betweenconsolidated and non-consolidated materials to separaterocks from other materials. For rocks, the highest leveldistinctions are made between aphanitic and phaneriticrocks and between rocks for which the origin can be deter-mined based on features visible in hand sample and rocksof indeterminate origin. Further refinement of the litho-logic nomenclature is based on five independent (orthogo-nal, unrelated) characteristics of the constituent parts,either individually or in combinations. These characteris-tics include grain size, grain shape, grain sorting, grain

GEOLOGIC CONCEPT MODELING, WITH EXAMPLES FOR LITHOLOGY AND OTHER GEOSCIENCE FEATURES 61

composition, and fabric (the relationship between thegrains). Examples of a lithologic type based on each ofthese characteristics are sandstone, breccia, porphyry,hornblendite, and schist. The description of a complexhand sample might include descriptions of several litho-logic components, each consisting of distinct assemblagesof minerals with distinct fabrics, and a set of relationshipsbetween these lithologic components.

Standard lithologic nomenclature is based first ondetermination of origin—igneous, sedimentary, or meta-morphic. In some cases, particularly for very fine-grainedor altered rocks, this determination requires some informa-tion about the context of the rock. The restriction to fea-tures visible in a hand sample requires descriptive termi-nology to allow for situations where the rock origin isindeterminate.

In more detail, the first order classifications in thedescriptive lithology system are:

1. Degree of consolidation (consolidated or non-con-solidated).

Consolidated is taken to mean that a piece of thematerial can be held in the hand as a single mass, and doesnot disintegrate into its constituent parts if struck with ahammer. The importance of this distinction is to separatematerials for which necessary and sufficient definition mayinclude criteria based on intergranular aggregate properties(e.g. sandstone) from those for which the definition mustbe based solely on the nature of the constituent particles(e.g. sand). Clearly there is a continuum from non-consol-idated to consolidated (e.g. gravel in an active stream toconglomerate, or laterite to underlying granite), but to beuseful for a computer database, a material must be one orthe other. A non-consolidated deposit may form an out-crop (e.g. a trench to expose a soil profile), and can thenbe described in terms of outcrop-scale lithology. For handsample lithologic description, a material is either a rock(consolidated) or a non-consolidated material.

2. Degree to which constituents can be discerned(aphanitic or phaneritic).

Rocks that have discernible constituents can bedescribed in terms of those constituents. If the material isaphanitic (constituents not discernible, i.e. diameter <~0.2mm), it is homogeneous for our purposes, and belongs tothe undifferentiated mineral matter and solid organic mat-ter types mentioned in the definition of rock. Phaneriticrocks are classified based on the nature of the parts thatmake the whole, and the relationship between the parts.

3. Genetic origin (sedimentary, igneous, metamor-phic, composite, anthropogenic, or indeterminate)

The third criteria separates material for which genericnames must be used from those for which the standardigneous, sedimentary, and metamorphic rock terminologycan be used. Rocks with features that demonstrate crystal-lization or cooling from a melted liquid are igneous.Rocks with features that demonstrate formation of the rockat low temperatures and pressures at or near the earth’ssurface are sedimentary. Conceptually, a metamorphicrock has a composite origin, but for consistency with stan-dard practice, metamorphic is considered to have the samerank as igneous, sedimentary, and anthropogenic. Thecomposite origin type includes rocks containing featuresindicative of a polygenetic origin that are not metamorphicrocks. The composite origin type includes various prob-lematic rock types. Volcaniclastic rocks are commonly notclearly of uniquely sedimentary or igneous origin.Composite rocks also include metasomatized (hydrother-mally altered) and weathered rocks that are differentenough from their original character to merit a new classi-fication, but are not typically thought of as metamorphic.Anthropogenic Earth materials are those that are the prod-uct of human activity. Since this classification scheme isdesigned to be descriptive, the criteria to distinguish thesevarious types must be based on features observable in ahand specimen. If no such features can be observed, therock origin is indeterminate.

This schema for defining lithology can be extended todescribe the Earth at larger scales. An outcrop descriptionconsists of a collection of lithologic components (hand-sample scale) and descriptions of the relationships betweenthe components. At the outcrop scale, the orientation ofgeometric elements of the fabric (in the Earth referenceframe) is introduced as a feature of a complete description.The terminology for this scale of observation probablycorresponds more closely to familiar geologic usage,because it allows for more genetic nomenclature based oncontext. A map-scale description consists of a collectionof map units (rock bodies) defined based on generalizationof outcrop-scale description, and descriptions of thenature, location, and orientation of boundaries between theunits (analogous to fabric on smaller scales). This descrip-tion schema can be applied to stratigraphic sections, for-mations, groups, supergroups, terranes....etc.

The boundaries that separate rock bodies from otherrock bodies are geological surfaces. A geologist studiesthe characteristics of these surfaces to infer the relation-ships between rock bodies. In detail, geologic surfaces arenormally boundary layers of finite thickness. For exam-ple, a fault might have a breccia zone of a particular thick-ness, and a gradational contact has a characteristic thick-ness between what is clearly rock A and clearly rock B.The boundary layer thickness limits the precision at whichthat surface can be located. At contacts between rock bod-ies, one rock will in general be older than the other.Exceptions include boundaries between metamorphic oralteration zones and facies boundaries. A fault surface has


an associated vector field that represents the slip on thefault at each point on the surface. Faults also have brack-eting ages for time of movement. Geologic surfaces thushave intrinsic properties of direction, thickness, and time.

A geologic map is a complex knowledge representa-tion. A standard geologic map depicts the location of rockbody boundaries as lines that bound closed polygons. Thepolygons are colored or labeled to indicate the geologicmap unit that is found at (or below!) the Earth’s surface atpoints within the polygon. The system used to classifyearth materials into the map units is described in accompa-nying text. The map units are often defined in terms ofgeologic formations that are only described in a cursoryfashion with reference to technical literature that providesmore complete information. On many maps the descrip-tion of the classification system is such that a user whodoes not have a strong background in geology would havedifficulty answering simple questions about the materialsthat might be expected to be associated with the outcrop ofthe map unit.

An important concept in analyzing the informationcontained in geologic maps is the distinction between adescription of a rock volume, and a description of a sur-face. Rock units describe the characteristics of a volumeof rock. Surficial geologic units describe the characteris-tics of the boundary layer between solid earth and atmos-phere or hydrosphere. Surficial units may describe thelithology of deposits to a depth that is small relative to thehorizontal extent of the map, or may relate to surface mor-phology, age (as opposed to deposit age), or depositionalenvironment. The colored polygons on a geologic mapeither represent the intersection of rock volumes with theearth’s surface (or map horizon), or represent characteris-tics of particular regions of the earth’s surface (or maphorizon). To a geologist interested in the processes andcharacteristics of the earth surface, the lines on the maprepresent boundaries of closed regions in the surface.Faults and rock-body contacts on a geologic map representthe intersection of two surfaces (rock body boundary andearth surface), projected onto a 2-D map surface. A geolo-gist interested in the rock bodies that compose the earthbeneath the map surface uses the 3-D geometry of theseintersection lines, along with measurements of surface ori-entations noted on the map, to understand the 3-D modelof the earth depicted by the map.

This 3-D analysis requires significant prior knowledgeof geologic map interpretation. Classification of contacttypes on geologic maps is generally binary-’fault’ or ‘con-tact’. Geologic reasoning must be used to determine therelationships between rock bodies. For example, consider-ation of rock type (e.g. granite), and rock age for adjacentrocks allows geologists to identify a contact as a con-formable depositional contact, nonconformity, angularunconformity, or intrusive contact. Accompanying textsometimes describes the nature of contacts that do not con-form to any simple classification. The geometry of faults

and contacts is either encoded on the map using a symbol(dip-direction arrow and dip magnitude), or it must beextracted by analyzing the location of intersection pointsbetween the contact trace and topographic contour lines.The geometry of faults and contacts can then be used todetermine underlying and overlying relationships notdescribed in the map legend. Once the geometrical rela-tionships of depositional and intrusive contacts are under-stood, the offset of intersection points of contacts at faultson the two sides of the fault for a series of contacts can beused to constrain possible magnitude and sense of dis-placement on a fault.

The classification of measurements of orientation offeatures (bedding, foliation, contacts...) shown on the mapis generally indicated by a set of symbols representingstandard feature types, with little or no explanation of thecriteria used to assign membership to a particular class. Inareas of complex structure, much of the orientation datacollected in the field is not shown on the map because ofspace limitations. The printed geologic map thus containsa great deal of information that is not explicitly stated, andprobably contains only a subset of the data and observa-tions made by the geologist in the field. This data can berecorded in a usable form in a computerized map database.

A geologic map is a visualization of the geologist’sknowledge about the spatial distribution of rocks and therelationships between them in a particular area. The stepsto produce this visualization are:

1. Select a map horizon.This is the surface that contains the physical features

to depict. It may be any arbitrary surface within the earth,but most commonly is the actual earth’s surface on whichgeologic observations are made.

2. Define the extent of the region in the map horizonto depict

3. Select a set of geologic entities to depict. Thesemay be rock volume or surface objects.

4. Determine the intersection of these volumes andsurfaces with the map horizon.

If the map horizon is non-planar (e.g. the Earth’s sur-face), this procedure will produce bounded 3-D surfacesand 3-D lines. A surficial geologic map is a special case inwhich the geologic surfaces to depict are identical with themap horizon, in this case the Earth’s surface. Other sortsof surface maps might show the distribution of fault rocksin a fault surface, or the distribution of alteration and min-eralization along a vein.

5. Project the 3-D surfaces and lines from earth coor-dinates on the map horizon to map coordinates in a planarsurface.


If the map horizon is planar (e.g. a cross section ormine-level map) this step only requires scaling. Thisresults in a map consisting of lines and polygons that rep-resent the intersection of the geologic entities of interestwith the map horizon, or the extent of geologic surfaceentities on the map horizon. These lines and polygons arecartographic entities.

6. Symbolize each cartographic entity.Assignment of graphical elements is based on the

classification of the geologic volume or surface entity rep-resented by the cartographic entity.

7. Superimpose the geologic map on a base map.The base map supplies the geographical context nec-

essary to relate the geologic relationships to the real world.If topographic information is included on the base map,the 3-D geometry of geologic lines on the map can bededuced to provide a basis for interpretation of the 3-Dgeometry of rock bodies.

One of the attractions of a digital geologic database isthe possibility of including a record of as much informa-tion as there is time or need to record about an area. Thegeologic data model must facilitate the storage, retrieval,and analysis of information at whatever level of detail it isavailable—whether the information is from detailed fieldnotes or a 50-year old reconnaissance map. System-spe-cific software tools based on the model would allow usersto produce derivative maps tailored to their needs, showingthings like distribution of rock containing more than 50%quartz, distribution of white rocks, distribution of fine-grained rocks, or all steeply dipping faults.

CONCEPTUAL MODELS OF SOME GEO-SCIENTIFIC ELEMENTS

In this section, some aspects of the earth sciencedomain described in the previous section are modeledusing the Object-Role Model (ORM) formalism (Halpin,1995). This formalism was chosen because it is expressiveenough to describe the necessary concepts, and techniquesare available to implement an ORM conceptual modelwith relational database software. Figure 1 summarizesthe notation and concepts used to graphically depict con-ceptual models presented below. The domain of discourseto be modeled must be separated into sub-domains tomake the graphical depiction more comprehensible.Models for two general purpose sub-domains are presentedfirst. These are Scalar Quantity and Fractional Analysis.The geologic sub-domain models included here areLithology Type, Particular Lithology, ConstituentGeometry, and Geologic Surface. These models assumethe existence of a standard classification table (analogousto the COA of Johnson et al., 1998) that contains defini-

tions of standard classification types. Entities referencedto this table are shown as values with a reference to‘ClassID’. All of these models are draft versions that arestill under development.

General Purpose Sub-domain Models

Quantities are most often used to represent measure-ments. They can be scalar (magnitude of magnetic field,isotopic date from a rock...), unit vectors (orientation ofsurfaces...), vector with magnitude (slip vector for fault,gravity or magnetic field...), or arrays (cobble counts, frac-ture orientation distribution...). These measurements servea wide variety of purposes, but are generally associatedwith a point or area with small dimensions relative to themap scale. The values obtained for these features ideallyare independent of the observer collecting the data.

Scalar Quantities (Figure 2) are used throughout themodel to specify numerical values. The Scalar Quantityconcept modeled here allows several representations. Thesimplest is ‘quantity has a single numeric value and a mea-surement unit attribute’. Possible more complex represen-tations include a single value with an uncertainty, or aminimum and maximum value with a specified defaultvalue to use if a single value representation is required. Inall cases the Scalar Quantity has a specified unit of mea-surement attribute. Subtypes of Scalar Quantity in otherFigures are identified by the use of QuantityID as the ref-erence mode for a subtype entity.

The Fractional Analysis (Figure 3) concept is used tomodel any situation in which an entity is composed ofother entities in certain proportions. Common examplesare chemical analyses, grain size-distributions, and modalmineralogy. Constituent types used in Fractional Analysesare specified in a system classification table. Constituentsthat are used as fractional parts of a described entity mustbe allowed as components of a Fractional Analysis of theDescribed Entity. If the quantity that specifies a fractionof some constituent is specified as a range ScalarQuantity, an algorithm is necessary to adjust default val-ues so that the constraint “sum of fractions = 1” can bemet. When a constituent is added in the analysis, a checkmust test that the sum of the minimum value for all thefractions is <= 1, and that the sum of the maximum valueis >= 1. These checks ensure that “sum of fractions = 1”is possible. If the sum of the fractions for constituent partsis <1, the difference is automatically classified as anunspecified constituent. Fractional Analysis entitiesappear in Particular Lithology schema (Figure 5) forChemical Composition, and in the Constituent Geometryschema for Grain Size Distribution.

Geologic Sub-domain Models

Two schemas are presented to model the concept oflithology. The General Lithology schema (Figure 4)


models the structure of a lithology type definition. Thesedefinitions provide the logical foundation for classificationof rocks into general lithologic classes. A geologic infor-mation system should contain a basic set of clearly definedlithologic terms in a standard classification table. This

schema closely follows the description of lithology in theDomain Analysis section (above). The General Lithologymay have widely varying degrees of specificity. The mostgeneral types are defined on only one aspect of the litholo-gy—e.g. sandstone, mudstone, schist.


An object-role model analyzes a domain in terms of elementary facts that assert that an entityhas an attribute or that one or more entities participate in an association. The relationshipbetween entities is defined by the roles in the association.

An association placed inside a box is 'objectified', andbecomes an entity that may play roles in other relationships.

entity

subtype1

isType

A subtype is linked with the parent type using athick line with arrowhead at the parent end. Theparent entity must be related to one and only onetype attribute that serves to identify the subtypes

subtype3subtype2

Constraints between roles are indicated by heavy dashed lines. In themodels here, the constraints are described by associated text.

Uniqueness constraint-an arrow spanning one or more roles indicates that the entityor combination of entities that play the spanned role(s) may appear only once in avalid database state.

Ternary relationship has three roles; uniquenessconstraints may include any combination of roles,but only most limiting constraints are indicated.

overliesis overlain

bysuperpositionrelationship

˚as, ˚ac

rock unit(rockUnitID)

Cyclical relationship--same entity plays both roles; ring constraints (˚xxabove connector between roles) are used to indicate kinds of relationsallowed. If R is the relationship, these can be stated logically as follows:˚ir--relationship is irreflexive. For all a, (a R a) is false˚as--asymmetric. For all a and b, if (a R b) then not (b R a).

Asymmetric relationships are always irreflexive˚ac--acyclic. There exists no path using the relationship from an entity

back to itself.

A simple reference scheme that associates each conceptual entity with exactly one value that represents theentity in the database is indicated by the name of the reference mode in parenthesis after the entity name.

object or entityvalue--number or text

stringhas applies to

binary association (two roles)

Roles

An association is symbolized by a string of connected boxes, one for each role in theassociation. A solid line connects the role box with the entity that plays the role.

An entity is representedby a solid ellipse

A value is representedby a dashed ellipse

Simple Attribute Relationship

Figure 1. Overview of Object-Role Model (ORM) graphical notation and conventions.

The Particular Lithology schema (Figure 5) modelsthe description of a particular rock. The information sys-tem should include a set of idealized ‘typical’ ParticularLithology entities associated with each GeneralLithology to provide default values for characteristics notspecified in the General Lithology definition. TheParticular Lithology entity will also be used for thedescription of rock hand samples. A Particular Lithologyis modeled as an aggregate of Constituents. TheConstituents may be Mineral or other Particular Lithology

entities, allowing for recursive description. EachConstituent may be involved in one or moreRelationship_within_Whole associations (see Table 1 forexamples) with the Particular Lithology. For each role aConstituent plays in the aggregate, a ConstituentGeometry can be described (Figure 6). This includesaspects such as grain size, grain shape, and sorting for aparticular Rock Constituent playing different roles, e.g. asgroundmass or as phenocrysts. Relationships betweenConstituents in particular roles within the whole aggregate


Numeric value(ValueID)

type

value type

{exact, uncertain}

value

uncertainty

value(number)

ScalarQuantity

(QuantityID)

type{value, range}

is greaterthan

is less than

has averageor default

quantity type

if a minimum value is specifiedfor a quantity, then a maximumvalue must also be present

uncertainty is required ifvalue type is uncertain

unit of measure

Units(ClassID)

Subtypes of Scalar Quantity throughout the modelspresented here are identified by the use ofQuantityID as their reference mode.

Figure 2. Schema for Scalar Quantity.

forms part of

composition association

Constituent(ConstituentID)

Fraction(quantityID)

FractionalAnalysis

(FractionalAnalysisID)

Analysis Type(classID)

describes iscan be part

ofmay

include

each FractionalAnalysis has a typeattribute used todistinguish subtypes

Subtypes of Fractional Analysis throughout the models presented here areidentified by the use of FractionalAnalysisID as their reference mode.

The sum of fractions for all constituentsin an Analysis must be 1.

a Constituent may be acomponent in morethan one Analysis Type

The Constituent that appears in a CompositionAssociation must be associated with theAnalysis Type for the Fractional Analysis inthe Composition Association.

Figure 3. Schema for Fractional Analysis.

are described by Relationship_between_constituent associ-ations (Table 2). These associations define the hand-sam-ple scale fabric and structure of the rock. A Mineral or aParticular Lithology may have physical property attribut-es such as color, density, magnetic susceptibility, soundvelocity, and refractive index. Chemical Composition is asubclass of Fractional Analysis (Figure 2), and may beassociated with a Mineral or a Particular Lithology.

Table 1. Examples of Relationship_within_Whole associ-ations

——————————————————————Forms matrix inDefines schistosityDefines graded bedsForms phenocrysts inForms porphyroblasts in

Table 2. Examples of Relationship_between_Constituentassociations:

——————————————————————ReplacesForms a rim onCrystallized beforeInterstitial toEnclosed in

Constituent Geometry (Figure 6) models a datastructure for describing the geometry of individual con-stituents in a Particular Lithology. These descriptionsare based on terms that may have qualitative or quantita-tive definitions (e.g. round, subround, euhedral, subhedral,fine-grained) or on free text descriptions referenced by‘descID’. The Constituent Geometry description has two


GeneralLithology

(ClassID)

mineralogy(FractionalAnalysisID)

grain size(FractionalAnalysisID)

grain sorting(DescID)

grain shape(DescID)

defines has

hasOrigin type

aggregate fabric(DescID)

defines has

defines has

defines has

defines has

{sedimentary, igneous, metamorphic, composite,anthropogenic, or indeterminate}

consolidation state

{consolidated, unconsolidated}

grain discernibility

has

has

{aphanitic, phaneritic}

Every General Lithology musthave one and only one value foreach of these attributes. A Rockmust have a consolidation stateof 'consolidated'

Every General Lithology musthave a value for at least oneof these attributes, and canhave at most one value forany one of these

Entities identified by DescID are text descriptions that specifycriteria; future models might add entities that define these informal or quantitative terms

Figure 4. Schema for General Lithology

sub-classes, Grain Geometry and Body Geometry, becausedifferent terminology is used when describing individualgrains or lithologic components in a mixed rock.

The Geologic Surface (Figure 7) schema requires thata Geologic Surface have a type attribute that determinesthe subclass membership for the surface, and an attributethat indicates if the surface is directed. A characteristicThickness of the surface may also be defined. Fault sur-faces have associated Slip Vectors that record displace-ments across the fault. A Fault may have several associat-ed slip vectors if displacement is known to have changedover time. Faults and Contacts both have associated AgeRanges. For a Fault, the age range is the time interval dur-

ing which the fault was active. For active faults, the lowerbound of this range is 0. Faults may have more than oneassociated age range to allow for more than one period ofactivity. Distinct age ranges may be related to distinct slipvectors to record the full displacement history of a Fault.The age range associated with Contacts represents the timeinterval that separates rocks on either side of the Contact.Only one age range may be associated with a Contact.Degenerate Volumes are entities used to represent descrip-tions of a discrete layer or a boundary layers that is toothin at the scale of representation to depict on a map.Examples include soil profiles, dikes, veins, fault rocks,and marker beds. A Degenerate Volume surface is associ-


Constituent

Mineral(MineralID)

relationship between constituents(ClassID)

relationship within whole(ClassID)

ParticularLithology

(plithID)

Fraction(QuantityID)

whole part

is composed of

... has relationship ... to ...

constituent geometry(ConstituentGeomID)

describes

Physical Property(ClassID)

ChemicalComposition(FractionalAnalysisID)

describes

describes

describes

describes

˚ac, ˚as

˚ir

is

is

forms part as

a Constituent is either a Mineral or another Particular Lithology; the relationship isequivalent to Mineral and Particular Lithology as subclasses that partition Constituent

a constituent may be present playing more thanone role in the aggregate (e.g. phenocrysts andgroundmass or clast and matrix); this relationshiprepresents each constituent/role combination, andthe constituent geometry is specific to thiscombination. The fraction of a constituent in thewhole is the sum of its fraction in all constituent/role combinations that include the constituent.

Typical grain to grain relationships or relationshipsbetween lithologic constituents. A constituent/rolecombination may not have a relationship to itself

Relationships are defined in a systemclassification table (COA of Johnson et al.,1998)

hasValue

(QuantityID)has

Physical property quantities are particular to particularconstituent/property associations. These propertydescriptions are independent of a particular lithologydescription. There may be several measurements for thesame property/ constituent pair

X

A Particular Lithology can not be aConstituent of itself. The constituent/particular lithology pair then hasfractional part and relationship to wholeattributes

a particular constituent has only onechemical composition

sum of all Fractions for a Particular Lithology must be 1 (see text)

Figure 5. Schema for Particular Lithology.

ated with a Rock Unit classification object that describesthe lithologic characteristics of the material in the bound-ary layer. A thickness must be defined for a DegenerateVolume surface.

CONCLUSIONS

This paper presents some simple conceptual modelsfor a small subset of the Earth Science domain of dis-course. Such models are an important precursor to imple-mentation of a large-scale database by formalizing the con-

cepts used to represent the domain of discourse. The mod-els presented here are by no means complete, and have notbeen tested against large sample data populations. Theprocess of developing these models has led to the recogni-tion of many unforseen conceptual difficulties, and has ledto the revelation of a compositional hierarchy data struc-ture that serves to describe many aspects of the domain.My own experience developing these models has con-vinced me of the importance of this step in the databasedesign process.


Constituent Geometry(ConstituentGeomID)

Geometry Type(classID)

has

GrainGeometry

BodyGeometry

describes

describes has

describes

describes

Sorting Term(ClassID)

Quantitativesorting description

(descID)

Grain shape term(ClassID)

Grain size term(ClassID)

Quantitative sizedistribution

(FractionalAnalysisID)has

has

has

describes

Thickness(QuantityID)

Lateral Extent(QuantityID)

Shapedescription

(descID)

geometry types are defined in systemclassification table (COA of Johnson et al., 1998)

A quantitive size distribution is a fractional analysis inwhich the constituents are size-range intervals

Values referenced by descID refer to free text descriptions.

Figure 6. Schema for Constituent Geometry.

APPENDIX A. GLOSSARY

Some definitions are presented to clarify the terminol-ogy used here.

Association: A concept that defines what a thing has to dowith one or more other things. An association has noindependent existence (Angus and Dziulka, 1998).Associations corresponds to logical predicates(Halpin, 1995).

Attribute: An inherent characteristic (Merriam-WebsterDictionary, 1999)

Class: A set of things sharing common attributes, definedby a set of criteria (a type). (Angus and Dziulka,1998).

Collection: A contingent aggregate of things, interpretedas an individual (Allgayer and Franconi, 1994).

Concept: A general idea, derived from specific instancesor occurrences (American Heritage Dictionary, 1982).An abstract or generic idea generalized from particular

instances (Merriam-Webster Dictionary, 1999)

Domain: The collection of things that may exist in a par-ticular model.

Entity: A thing that has independent, separate, or self-con-tained existence (Merriam-Webster Dictionary, 1999).An instance of a type that has its own identity and canbe distinguished from another instance that meets thedefinition of the same type. This implies that an enti-ty has an additional property that is its unique identifi-er.

Instance: A thing that is representative of a type(American Heritage Dictionary, 1982). An instancedoes not have a unique identity (compare to entity);two instances representative of the same type can notbe distinguished (Angus and Dziulka, 1998).

Individual: A single thing in a domain (Allgayer andFranconi, 1994)

Primitive Concept: A concept that has no internal struc-ture or cannot be defined in terms of necessary andsufficient properties (MacRandal, 1988).


Geologic Surface(SurfaceID)

is

is directed

type(ClassID)

has direction

Slip vector

age range(quantityID)

slipped in thisdirection during

has time between juxtaposedunits

hasthickness

(quantityID)

is

Rock Body type(RockUnitID)

A fault with slip vectordefined is always directed

Surface types are defined in a database classificationtable (COA of Johnson et al, 1998)

Rock Body types are defined in adatabase classification table(COA of Johnson et al, 1998),referenced by RockUnitID

Faults may have more than one period ofmovement, with different slip vectors defined foreach period. Slip vectors and ages may be knownfor only some of the periods of movement. Theage range for different periods of movement mustbe non-overlapping. If a fault is involved in a 'hasdirection' relationship that has an associated agerange, it must also be involved in a 'was/is activeduring' relationship for the same age range.

Uniqueness constraint indicates a particularcontact may represent only one age rangebetween the units juxtaposed at the contact.The age range could be a derived value,based on the ages of rock bodies adjacent tothe contact.

was/is active during

Fault

Contact DegenerateVolume

The Typeattributeserves todifferentiatesubclasses

The slip vector value is not defined yet; it is aVector Quantity or Vector Field , dependingon whether slip is defined at one point or manypoints on the fault surface.

Thickness must bedefined for DegenerateVolume surfaces

Figure 7. Schema for Geologic Surface.

Relationship: A connection between an association and athing. The relationship is identified (with respect tothe association) by the role played by the related thingin the association (Angus and Dziulka, 1998).

Role: The thing that receives or is affected by the actionof an association. A concept that indicates in whatcapacity some thing is involved with some otherthing. Roles are used in naming relationships withrespect to an association (Angus and Dziulka, 1998).Roles hold information about function.

Specific Thing: An individual occurrence of some thing;an entity (Angus and Dziulka, 1998).

Taxonomy: A superclass/class/subclass hierarchy in whichevery instance of a child class is a member of theclass(es) that are parent of the child.

Thing: An entity, situation, association, or event that maybe perceived, known, or imagined.

Type: The semantic distinctions between ‘type’, ‘class’,and ‘domain’ are ambiguous and inconsistentlyapplied in the literature. Type here will be taken toapply to the set of attributes that determine, for anything, whether that thing is or is not a member of aclass. A type defines a domain; a class is a set ofthings that belong to the domain. The domain is theabstract collection of all things that meet the definitionof the type. Instances of types do not have an identity,i.e., two instances of the same type can not be distin-guished (definition distilled from many sources).

Typical Thing: A type based on a particular set of entities,characterized by a subset of the attributes of the realworld entity(ies), and/or idealizations of those attribut-es. Attribute values are assigned, not measured, andare normally specified as a possible range of values(Angus and Dziulka, 1998).

APPENDIX B. DATA MODELING

This section is a review of data modeling literatureand activities that provide a context for efforts to develop aUSGS/AASG geologic map data standard (Johnson et al.,1998). In order to communicate knowledge about a partic-ular domain, there must be some agreement on the con-cepts and roles defined in that domain. This commonframework of concepts and roles is called a conceptualmodel for the domain in question. Much research intoconceptual models, or knowledge representation, has takenplace in the artificial intelligence arena. These efforts areimportant to geoscience data modeling because theyattempt to formalize systems for describing the real world.Such systems are necessary to consistently map geologic

knowledge into logical models that can be implemented incomputer databases. The most common approach to mod-eling descriptive knowledge of the sort recorded on geo-logic maps is based on the idea of a structured inheritancesemantic network (Brachman and Schmolze, 1985;Ringland and Duce, 1988). There are other models forknowledge systems, but this approach is used here becauseit lends itself well to implementation with current comput-er systems, and corresponds well to the thought processesused by geologists.

A semantic network is formed by defining a set ofprimitive concepts, or abstractions of fundamental thingsin the real world. Concepts are the things we can talkabout, the epistemological primitives of our thoughtprocess. A primitive concept commonly corresponds to a‘natural kind’—something abstracted from the real world,that cannot be defined logically in terms of necessary andsufficient conditions. A partial definition of a primitiveconcept can include necessary conditions. Derived con-cepts are defined by logical combination of concepts, orassociation of a concept with attributes and constraints toidentify a subset of the concept. An association is anordered pair of concepts and a predicate that defines therole of the association. The associations between the partsof a derived concept define the structure of the thing beingmodeled. A constraint is an association between parts of aderived concept that has a boolean value property whosevalue depends on properties of the instances of things itinvolves. For an instance of a concept to be valid, all itsconstraints must evaluate to true. In a semantic network,the concepts are represented as nodes, and the associationsas connections between the nodes.

There is widespread interest in methods for develop-ing and expressing data models, driven by efforts to auto-mate various business operations. Recent contributionsinclude XML-Data (<http://www.w3.org/TR/1998/NOTE-XML-data>), which is an extension to the ExtensibleMarkup Language (XML) (<http://www.w3.org/TR/REC-xml>). XML-Data was designed to communicate informa-tion about the structure of data over the world wide web.This will allow applications to be produced that couldacquire and analyze data sets from a variety of sourceswithout prior knowledge of the data structure underlyingthe data set. The Universal Modeling Language (UML)(<http://www.rational.com/uml>) is a system for modelingthe structure of data and operations necessary to utilize thedata for a particular purpose. This language was motivat-ed by efforts to provide a standard mechanism for describ-ing complex data management operations, thereby facili-tating the development of computer-aided software engi-neering (CASE) tools. UML is strongly based in Object-oriented system thinking. For a comparison of UML withthe Object-Role Modeling approach used here see Halpin(1998).


Another important effort to develop tools for datamodeling has been driven by the InternationalOrganization for Standardization (ISO) project 10303,Industrial automation systems and integration, Productdata representation and exchange (referred to as theInternational STandard for the Exchange of Product Dataor STEP). As part of this project, a formal informationrequirements specification language, named EXPRESS,has been developed (Spiby, 1998). EXPRESS and UMLappear to provide similar modeling capabilities.

In connection with the ISO 10303 project, a groupnamed EPISTLE (European Process Industries STEPTechnical Liaison Executive) has published a documenttitled “Developing High Quality Data Models” (West,1996). This document discusses the rationale for a sys-tematic approach to data models, and lays out a set of prin-ciples for developing data models that are stable, flexibleto changing practices, and extensible to changing needs.These principles include:* Candidate attributes should be treated as representing

relationships to other entities

* Associations should be represented by entity types (notrelationships or attributes)

* Relationships (in the entity/relationship sense) shouldonly be used to express the involvement of entitytypes with associations.

* Entity types should represent, and be named after, theunderlying nature of an object, not the role it plays ina particular context.

* Entity types should be part of a subtype/supertype hier-archy that defines a universal context for the model.The context is the range within which the model isvalid. A data model also has a scope that encompass-es what is actually modeled. If the context for themodel is universal, then any extension to the scopewill always be contained within the context.

The EPISTLE group has also published a Frameworkdocument (Angus, and Dziulka, 1998), that provides ahigh-level (meta) model and a set of core constructs thatserve as a basis for deriving ‘high-quality’ conceptualmodels. I have attempted to incorporate the EPISTLEframework into the models presented here.

APPENDIX C. SURVEY OF RELEVANTDATA MODELS

Spatial Data Model

The OpenGIS consortium is attempting to develop aset of specifications to allow ‘interoperable geoprocessing’

(Open GIS Consortium Technical Committee, 1998). Partof this effort is a specification for conceptually represent-ing the Earth and Earth phenomena. Discussions of thenature of the “National Geologic Map Database”(<http://ncgmp.usgs.gov/ngmdbproject>) commonlyinclude consideration of the database as a set of resourcesdistributed over separate nodes that may be operating indifferent software/hardware environments (Soller andBerg, 1997). The OpenGIS specification is being designedas a means of implementing exactly this sort of distributedspatial database. To the extent that this specification gainsacceptance and software becomes available to implementthe services required, this effort could provide the frame-work for developing a National Geologic Map Databasethat would conform to an industry standard. Conformancewith such a standard would mean that the geologic com-munity would not have to develop a lot of application-spe-cific software.

Significant development of spatial data models in thespecific context of earth science has also been necessary inthe development of software applications for visualizationof 3-D geologic models. The GOCad Research Programwas initiated in 1989 by the Computer Science group ofthe National School of Geology (ENSG) in Nancy, France.The goal of this project is to develop a new computer-aided approach for the modeling of geological objectsspecifically adapted to Geophysical, Geological andReservoir Engineering applications (c.f. Royer et al.,1996). The project is supported by an international con-sortium of oil companies and academic institutions. [Formore information see web page at <http://www.ensg.u-nancy.fr/ GOCAD/ Welcome.html> and<http://www.ensg.u-nancy.fr/GOCAD/doc/ref_man.html>.] The documentation on the web sitereports compliance with the POSC standard (see below).3DMove (<http://www.mve.com/p-3dmove.html>) is asimilar commercial software package developed for visual-ization of 3-D geologic structure, but no information isavailable on the underlying data model.

Geoscience Data Models

A great deal of effort has been made by the petroleumindustry to develop a logical data model to facilitate theinterchange of data. Two major initiatives have proposedmodels: Petrotechnical Open Software Corporation(POSC), and the Public Petroleum Data Model (PPDM).The POSC model is based to a large extent on the frame-work of ISO 10303 (STEP). Its scope includes technicaland business aspects of petroleum exploration, develop-ment, and production (POSC, 1997). The complexity ofthe model is daunting, and the documentation providesonly the basic information necessary to define the model.Study of the model suggests to me that it may provide


nearly all the data constructs necessary for the purposes ofthe National Geologic Map Database, but more detailedanalysis is necessary to determine if the conceptual modelsproposed here can be mapped into constructs in the POSCmodel.

Some problems with the POSC model have been iden-tified. First, the model introduces ‘named data types’ thatare outside of the data types defined in the EXPRESS lan-guage specification (Spiby, 1998). Thus, public domainsoftware designed to interpret and edit conceptual modelswritten in EXPRESS does not work with the POSC model.Second, the scope of earth science included in the model islimited to aspects relevant to petroleum exploration andproduction. There are no provisions for the complexitiesof igneous and metamorphic rocks, or for describing surfi-cial geology, beyond those that overlap with the geologyof sedimentary rocks.

The Public Petroleum Data Model is the second datamodel developed to standardize data modeling by thepetroleum industry. Details of this model are availableonly to members of the association, and have not beenstudied by the author. More information is available at<http://www.ppdm.org/>.

A number of government geological surveys havedeveloped models for internal use. These are documentedto varying degrees. Workers from the British GeologicalSurvey (BGS) have published a series of papers on geo-logic map database models, data dictionaries, and stan-dardization of mapping practices (Laxton and Becken,1996; Bain and Giles, 1997; Giles, 1997; Allen, 1997).The paper by Bain and Giles (1997) outlines the frame-work of the data model used by BGS using entity-relation-ship diagrams, but does not give detailed definitions of theentities used. This paper points out the importance ofdefining the mapping horizon (GEOLOGICAL LEVEL intheir terms) as a proxy for elevation data associated withx,y points on a geologic map.

The Australian Geological Survey Organisation(AGSO) has been involved with geological databasedevelopment since the 1970’s, and is a participant in thePOSC and PPDM efforts. Their system has apparentlydeveloped from the ground up, evolving into an integrateddatabase system with the OZROX Field Geology Databaseat its center (Ryburn et al., 1995). OZROX contains infor-mation on field locations, outcrop data, measured sectionand drill hole logs, lithology and sample data, and structur-al observations. It is linked with databases for geochronol-ogy (OZCHRON), whole-rock geochemistry(ROCKCHEM) petrography (PETROGRAPHY), bios-tratigraphy (STRATDATA), the Australian national petro-leum database (PEDIN), a bibliographic database forsource citations (AGSOREFS), and standard referencedatabases for Stratigraphic Names, Geological Provincesand the Geological Time Scale. A published description of

the conceptual model underlying OZROX has not beenfound.

AGSO arranged for the Australian Mineral IndustriesResearch Association (AMIRA)(<http://www.amira.com.au/>) to contract the developmentof a standard geoscience data model across the mineralexploration industry. The AMIRA project P431,“Geoscience Data Model” was delivered in April, 1998(Ryburn and O’Donnell, 1998), consisting of entity-rela-tionship diagrams and a data dictionary. The model isdescribed in Miller et al. (1998). It models the domainsof geology, geochemistry, drilling, and mineral resourceinformation. The model is designed to provide a standarddatabase framework for mineral exploration companies tofile required reports on their exploration activities withgovernment agencies, and to facilitate data exchangebetween companies. It provides elegant data structures forstoring information on boreholes, rock samples, mineraldeposits (orebody geometry, commodities, production),geochemical analyses, and relationships between compa-nies, exploration projects, and tenements (mineral leases).The model includes a ‘profile’ construct to provide meta-data defining customized implementations of the fullmodel. A conceptually similar ‘template’ construct pro-vides metadata to describe sets of chemical analyses.Detailed geologic descriptions are included through fieldsfor stratigraphic unit, rock type, lithological name, andlithological descriptor and text comments that may beapplied to any ‘fraction’ of a mineral deposit, bore hole(bore hole interval), rock outcrop, or individual sample.Fabric elements may be related to any of these rock enti-ties, and may have relationships defined with other fabricelements. Almost any entity in the model may have one ormore free text ‘observations’ associated with it. Themodel relegates development of systems of terminology toindividual implementations of the model. The model doesnot attempt to describe geologic maps.

The Commonwealth Scientific and Industrial ResearchOrganisation (CSIRO) in Australia has also developed amodel for Geoscientific Spatial Information systems(Lamb et al., 1996; Lamb et al., 1994; Power et al., 1995;<http://www.ned.dem.csiro.au/research/visualisation/DMGE/>). This is presented as an object-oriented data modelthat contains classes describing: geometry and topology of3D geoscience entities, an audit trail for tracking sourcesof information, an information quality hierarchy, a spa-tiotemporal database record hierarchy, and a few geoscien-tific entities. This is a physical model for implementationin a C++ programming environment. CSIRO is a memberof the GOCad consortium and their model apparently usesa framework similar to that used by GOCad for spatial andtopological data.

The Geological Survey of Canada (GSC) has support-ed development of the FieldLog software system(<http://gis.nrcan.gc.ca/fieldlog/Fieldlog.html>) to aid


geologist manage geologic field data. The data modelunderlying FieldLog is described in general terms inBrodaric (1997). This model has been used for a numberof years by geologist for collection and compilation offield data, and appears to provide a flexible, robust, andexpressive structure for collecting, archiving and analyzingthe geologic data that support spatial data presented ongeologic maps. The GSC has also published a cartograph-ic database standard that is a physical model for geologicmaps in an Arc/Info environment(<http://www.nrcan.gc.ca/ess/carto/english/reference/GSCCDBS.pdf>).

The Geologic Survey of Colombia (INGEOMINAS)has published some information on their model for geo-science data (Murillo, 1995). This data model wasdesigned for INGEOMINAS as a means to integrate data-base operations over the domains of geology, geophysics,mining, geoenvironmental engineering, samples and wells.The fundamental entities in this data model areObservation Point (any location), the Spatial ReferencePlane (spatial coordinate frame for Colombia), and theMapping Terrain Unit (a closed surface object having char-acteristics different from surrounding units). No underly-ing conceptual model is elucidated in the description ofthis model.

Most participants at this conference are probablyfamiliar with the standard data model proposed by theUSGS, AASG and GSC (Johnson et al, 1998). Otherefforts in the USGS have produced implicit or explicit datamodels. Implicit models underlie two significant softwareprograms: AlaCarte, developed for geologic map dataentry using Arc/Info (Fitzgibbon, 1991; Wentworth, 1991),and GSMCAD (Williams et al., 1996), developed for PC-based geologic mapping. Descriptions of the conceptualmodels underlying these packages are buried in the detailsof the physical implementations. Matti et al. (1997a,1997b, 1997c) describe a unique physical data model usinga linguistic root-suffix coding scheme. This model pro-vides extensive hierarchical word lists that serves as anexcellent guide for the concepts that need to be represent-ed in a successful geologic knowledge base.

Acknowledgements. Thanks to Claire Zucker (PimaAssociation of Governments) for a careful editorial review.My thinking has been shaped by instructive and stimulat-ing conversations with Boyan Brodaric (GSC), GaryRaines, Bruce Johnson, Ron Wahl, Jon Matti (all USGS),and Tim Orr (AZGS). John Tsotsos (University ofToronto) kindly provided a copy of an unpublished manu-script that introduced me to the engineering/artificial intel-ligence approach to geoscience knowledge representation.

REFERENCES

Ady, B.E., 1993, Towards a theory of spatio-chronologic rela-tions for geoscience: Sudbury, Ontario Geological SurveyOpen-File Report 5854, 76 p.

Allen, P. M., 1997, Standardization of mapping practices in theBritish Geological Survey: Computers in Geosciences, Vol.23, p. 609-612.

Allgayer, Jurgen, and Franconi, Enrico, 1994, Collective entitiesand relations in concept languages, in Lakemeyer, Gerhard,and Nebel, Bernhard, eds., foundation of KnowledgeRepresentation and Reasoning: Berlin, Springer-Verlag, p.13-29.

American Heritage Dictionary, 1982, Second College Edition:Boston, Houghton Mifflin Co., 1568 p.

Angus, Chris, and Dziulka, Peter, 1998, EPISTLE FrameworkV2.0, Issue 1.22. Available at<http://www.stepcom.ncl.ac.uk/epistle/data/framedoc.htm>;or contact current editor: Peter Dziulka, Keyworth Institute,Department of Mechanical Engineering, University ofLeeds, Woodhouse Lane, Leeds LS2 9JT.

Bain, K.A., and Giles, J.R.A., 1997, A standard model for storageof geological map data: Computers & Geosciences, v. 23,no. 6, p. 613-620.

Bates, R.L., and Jackson, J.J., eds., 1987, Glossary of Geology,Third Edition: Alexandria, VA, American Geologic Institute,788 p.

Bernknopf, R.L., Brookshire, D.S., Soller, D.R., McKee, M.J.,Sutter, J.F., Matti, J.C., and Campbell, R.H., 1993, SocietalValue of Geologic Maps: Washington, D.C., U.S.Geological Survey Circular 1111, 53 p.

Brachman, R.J., and Schmolze, J.G., 1985, An overview of theKL-ONE Knowledge representation system: CognitiveScience, v. 9, p. 171-216.

Brodaric, Boyan, 1997, Field data capture and manipulationusing GSC Fieldlog 3.0, in Soller, D.R., ed., Proceedings ofa workshop on digital mapping techniques: Methods forgeologic map data capture, management, and publication:U.S. Geological Survey Open-File Report 97-269, p. 77-82.

Fitzgibbon, T.T., 1991, ALACARTE Installation and SystemManual, Version 1.0: U.S. Geological Survey Open-FileReport 91-587B. Available at<http://wrgis.wr.usgs.gov/docs/software/software.html>.

Giles, J.R.A., Lowe, D.J., and Bain, K.A., 1997, GeologicalDictionaries—Critical elements of every geological data-base: Computers & Geosciences, v. 23, no. 6, pp. 621-626.

Halpin, T.A., 1995, Conceptual Schema and Relational DatabaseDesign, Second Edition, Prentice Hall Australia, 547 p.

Halpin, T.A., 1998, UML data models from an ORM Perspective:Part One: Journal of Conceptual Modeling,<http://www.inconcept.com/JCM/April1998/print/halpin.html>.

Johnson, B.R., Brodaric, Boyan, and Raines, G.L., 1998, DigitalGeologic Maps Data Model, v. 4.2: AASG/USGS DataModel Working Group Report. Available at<http://ncgmp.usgs.gov/ngmdbproject/ standards/datamodel/model42.pdf>.

Lamb, Peter, Horowitz, Frank, and Schmidt, H.W., 1996, A DataModel for Geoscientific Spatial Information Systems, ver-sion 1.2, Draft Nov. 27, 1996: CSIRO Division ofInformation Technology. Available at<http://www.ned.dem.csiro.au/ research/visualisation/DMGE/specs/paper.htm>.


Lamb, Peter, Horowitz, Frank, and Schmidt, H.W., 1994. A datamodel for geoscientific spatial information systems: Version1.0: Canberra, Australia, CSIRO Division of InformationTechnology, Technical Report TR-HJ-94-05.

Laxton, J.L., and Becken, K., 1996, The design and implementa-tion of a spatial database for the production of geologicalmaps: Computers and Geosciences, v. 22, p. 723-733.

MacRandal, Damian, 1988, Semantic Networks, in Ringland,G.A., and Duce, D.A., eds., Approaches to KnowledgeRepresentation: an Introduction: New York, John Wiley andSons, Inc, p. 45-79.

Matti, J.C., Miller, F.K., Powell, R.E., Kennedy, S.A.,Bunyapanasarn, T.P., Koukladas, Catherine, Hauser, R.M.,and Cosette, P.M., 1997a, Geologic-point attributes for digi-tal geologic-map databases produced by the SouthernCalifornia Areal Mapping Project (SCAMP), Version 1.0:U.S. Geological Survey Open-File Report 97-859, 37 p.

Matti, J.C., Miller, F.K., Powell, R.E., Kennedy, S.A., andCosette, P.M., 1997b, Geologic-polygon attributes for digi-tal geologic-map data bases produced by the SouthernCalifornia Areal Mapping Project (SCAMP), Version 1.0:U.S. Geological Survey Open-File Report 97-860, 42 p.

Matti, J.C., Powell, R.E., Miller, F.K., Kennedy, S.A., Ruppert,K.R., Morton, G.L., and Cosette, P.M., 1997c, Geologic-lineattributes for digital geologic-map databases produced bythe Southern California Areal Mapping Project(SCAMP),Version 1.0: U.S. Geological Survey Open-FileReport 97-861, 103 p.

Merriam-Webster Dictionary, 1999, <http://www.m-w.com/dictionary.htm>.

Miller, D.R., Hume, R.G., and Parker, A.J., 1998, TheGeoscience Data model, AMIRA Project P431 Final Reportto Sponsors: Australasian Spatial Data Exchange Centre,ref.#12587.

Murillo, Alvaro, 1995, A GIS Data Model Prototype: ESRIAnnual User’s Conference Proceedings. Available at:<http:// www.esri.com/library/userconf/proc95/to050/p017.html>.

Open GIS Consortium Technical Committee, 1998, The OpenGISGuide, Third Edition, edited by Kurt Buehler and LanceMcKee: Open GIS Consortium, Wayland, Massachusetts,USA. Available at<http://www.opengis.org/techno/guide.htm>.

POSC, 1997, POSC Specifications, Version 2.2: Houston, Texas,Petroleum Open Software Corporation, 1 CD ROM.Available at <http://www.posc.org>.

Power, W.L., Lamb, Peter, and Horowitz, F.G., 1995, Data trans-fer standards and data structures for 3D GeologicalModelling, in Application of Computers and OperationsResearch in the Minerals Industries: Carlton, Victoria,Australia, Australasian Institute of Mining and MetallurgyPublications Series No. 4/95, ISBN 1 875776 26 5.Available at <http:// www.ned.dem.csiro.au/ research/visualisation/papers/power.html>.

Ringland, G.A., and Duce, D.A., editors, 1988, Approaches toKnowledge Representation: An Introduction: New York,John Wiley and Sons, Inc., 260 p.

Royer, Jean-Jaques, Gerard, Benoit, LeCarlier de Veslud,Christian, and Shtuka, Arben, 1996, 3D Modeling ofComplex Natural Objects, in Dubois, J.-E., and Gershon,N., eds, Modeling Complex Data for Creating Information:Berlin, Springer, ISBN 3-540-61069-3, p. 155-168.

Ryburn, Rod, Bond, Lynton, and Hazell, Murray, 1995, Guide toOZROX - AGSO’s field geology database, AustralianGeological Survey Organisation Record 1995/79. See also<http://www.agso.gov.au/information/structure/ isd/database/ozrox.html>.

Ryburn, Rod, and O’Donnell, Ian, 1998, Towards national geo-science data standards: AGSO Research Newsletter v. 29,Nov. 1998.

Soller, D.R., and Berg, T.M., 1997, The National Geologic MapDatabase - A Progress Report: Geotimes, v. 42, no. 2.Available at <http://ncgmp.usgs.gov/ngmdbproject/geot-imes97.html>.

Spiby, P., ed., 1998, Production data representation and exchange,Description methods, The EXPRESS language referencemanual: ISO Document TC184/SC4/WG11 N48, 296 p. ,<http://www.nist.gov/sc4/wg_qc/wg11/n047/wg11n047.zip>

Wentworth, C.M., 1991, ALACARTE User Manual, Version 1.0:U.S. Geological Survey Open-File Report 91-587C.Available at <http://wrgis.wr.usgs.gov/docs/software/software.html>.

West, M., 1996, Developing High Quality Data Models, Issue2.1, EPISTLE. Available at<http://www.stepcom.ncl.ac.uk/epistle/data/mdlgdocs.htm>;or from Matthew West, Shell International Ltd., ISCL/4Shell Centre, London, SE1 7NA, UK.

Williams, V.S., Selner, G.I., and Taylor R.B., 1996, GSMCAD, anew computer program that combines the functions of theGSMAP and GSMEDIT programs and is compatible withMicrosoft Windows and Arc/Info: U.S. Geological SurveyOpen-File Report 96-0007, 18 p., one 3.5-inch disk.Available at <ftp://greenwood.cr.usgs.gov/pub/open-file-reports/ofr-96-0007/>.


77

INTRODUCTION

After more than ten years of application of digital sys-tems to the compilation, editing, and publication of moder-ate to large-scale geologic maps, USGS geologists on theWestern Geologic Mapping Team routinely produce, andmany of our customers expect delivery of, both digital spa-tial databases and paper map plots instead of traditionalprinted maps. Efficiencies in publication are also urgingus towards release in digital form. We are beginning toencounter problems, however, because of differences ingeology, applications, customer expectations, and the lackof agreement on standards for large-scale geologic mapdata. The very energy and creativity of the digital geolog-ic mappers themselves have led to a proliferation ofapproaches to the preparation and delivery of modern geo-logic maps.

Within the Western Geologic Mapping Team, severalfunded projects have developed different approaches todigital geologic-map development. In our Las Vegas pro-ject, maps have been produced using a CAD program andthen converted into Arc/Info format for release (for exam-ple, McKee and others, 1998). In the San Francisco Bayproject, scientists ordinarily use ALACARTE (Fitzgibbon,1991; Fitzgibbon, and Wentworth, 1991; Wentworth, andFitzgibbon, 1991), a menu and standards system forArc/Info developed ten years ago as one of the first suc-cessful systems for rendering geologic map informationdigital (used, for example, in Brabb and others, 1998). Insouthern California, the Southern California ArealMapping Project has developed a more complex datamodel (Matti and others, 1997a,b,c) to produce numerousmaps for publication and for fulfillment of various con-tracts (for example, Miller and others, 1998), while in thePacific Northwest, geologic mappers hope for betterweather and input their data in their own fashion.

If such variation exists within just one team, considerthe diversity of approaches across the whole spectrum ofproducers and users of geologic maps. And yet the needfor effective communication with customers and internalefficiency should be driving us towards common practiceand standards. We have an urgent need to present a com-mon face to the outside world and to be able to move datasets from one place or project to another. This paperdescribes my effort to move a group of creative and pro-ductive geologists towards agreement in the form of theirdigital products without interrupting the flow of work ordenying needed flexibility in content and procedures.

GOING ENTIRELY DIGITAL

Two years ago, I began a campaign to expand digitalwork in the team to ensure that all newly released mapspublished by the Western Geologic Mapping Team werefully digital, including a graphics file (the cartographicmap), a readme (first cousin to metadata), and a spatialdatabase. Many team members were already working thisway, but for some this required moving from digital graph-ics to GIS, and for others it required diving into the digitalworld for the first time.

In implementing this digital-only policy, it soonbecame clear that, as is often said, the devil is in thedetails. The serious practical issues surrounding databasedesign, content of layers, needed attributes and the codesto describe them, the role of relates and lookup tables, anda myriad of other details presented themselves to the geol-ogists immediately. Work simply couldn’t proceed withoutpractical means of addressing these seemingly mundaneissues. At the same time, demanding users expected, byturns, traditional printed materials, preliminary plots,Postscript and PDF files, or full-blown ARC coverages to

Data Model for Single Geologic Maps: An Application of theNational Geologic Map Data Model

By Donald L. Gautier

U.S. Geological Survey345 Middlefield Road, MS 975

Menlo Park, CA 94025Telephone: (650) 329-4909

Fax: (650) 329-4936E-mail: [email protected]


be downloaded in their own computers, all with rapiddelivery. The various projects, with their resourceful sci-entists, seemed to find almost innumerable ways to solvethese problems and deliver the required materials. Thiscreative diversity came, however, at the price of consisten-cy, compatibility, and efficiency.

Fundamental to a digital geologic map is the organiza-tion of its spatial data into topical layers and databasetables, as well as the encoding of its other parts (table 1).It became clear that we all needed to agree on a standardmodel for encoding the spatial data before we could pro-ceed to address the other details. Within the team we hadthe data model inherent in ALACARTE (the ALC datamodel), the new and rapidly evolving SCAMP data model,as well as other developing and ad hoc data models not yetformally documented.

It was also clear that, regardless of any future migra-tion to other applications, we now needed to work withinthe practical constraints of Arc/Info: the Geologic Divisionrequires that map datasets be released in Arc/Info as wellas SDTS format (GD Policy 6.1.3.2), Arc/Info has becomevirtually the lingua franca of geologic mapmakers, andARC export format is the single most useful exchange for-mat for our customers.

At the national level, the draft National Geologic Mapdata model (Johnson and others, 1998a, b) provides astructure for a national data library that is designed toaccept data from, and track the differences between, manydisparate geologic maps. Accommodating these differ-ences leads to complexity that is not needed for individualgeologic maps, and the model cannot be implemented inArc/Info. It does, however, provide us with a usefulframework to help guide our development of team stan-dards for the release of new, medium- to large-scale mapsproduced within the context of a team mapping project.

We call this partial implementation of the NGM datamodel the Single Geologic Map (SGM) data model. Bysingle we mean that the model addresses the encoding ofindividual geologic maps. The SGM data model must beable to encode all the information represented on each mapin such a way that the digital version of the map (includingboth the spatial and attribute data) can be released as aself-contained USGS publication. Adjacent maps withcompatible stratigraphies, line categorizations, etc., can bemosaicked to create a regional data library or the maps canbe uploaded into the national database.

GOALS AND CONSENSUS

The single geologic-map data model is being devel-oped under three broadly defined objectives: (1) to contin-ue to serve the needs of the map users, (2) to provide con-sistency and compatibility within the team, and (3) to becompatible with the single-map components availablewithin the NGM data model.

The SGM data model must have certain properties. Itneeds to address the encoding of the information foundwithin a standard paper geologic map (table 1), and itneeds to define an agreed-upon model within which thedata can be stored and exchanged. The data model mustbe backward compatible to the extent that all our previouscreative and labor-intensive development work is not wast-ed. It must be as simple as possible and it must be able tobe implemented in Arc/Info.

In developing this model, it has been important to fol-low a process that could develop consensus among themap producers of the team. The model has to build uponthe current state of knowledge and application, thus takingadvantage of our existing experience and capability andavoiding the need for complete retooling. It has to draw inthe existing practitioners by providing them clearly recog-nizable advantages, and it has to be able to accommodatenew requirements while retaining maximum flexibility(future expansions of the model shouldn’t break earlierdatasets). Finally, we have had to create a model that isnot too great a departure from the products with which ourusers are already familiar and able to use.

THE SINGLE GEOLOGIC MAP (SGM)DATA MODEL

The SGM data model defines two fundamental geo-logic map layers: a units, contacts and faults (UCF or geol-ogy) layer, and a structural data (SD) layer. Other layersare permitted but not defined at this time. SGM uses thevector spatial data model, also known as the geo-relationalor arc-node-topological model, in which points, lines andpolygons (area boundaries) represent spatial features.Attributes are stored in relational database tables linked tothe spatial features by common identification numbers.Standard vocabularies for attributes are not yet defined inthe SGM model, but all attributes must be documentedwithin the dataset’s definition tables.

Table 1. Typical parts of a standard geologic map

Map graphic Spatial data, drawn in a projection in XY spaceDescription of map units DMU: map symbols and text descriptions of the unitsCorrelation of map units CMU: graphic depiction of the age relations of map unitsTextCross section(s) Spatial data, drawn along a vertical profile (in XYZ space)Other illustrations

Each of the two map layers consists of a spatial layerand its associated attribute tables (figure 1). Two types ofspatial features are grouped in each layer: arcs (faults andcontacts) and polygons (geologic units) in the UCF layer,and arcs (structural lines) and points (structural measure-ments) in the SD layer. This natural grouping facilitatesqueries and avoids the problems of duplicate storage. Itcan be implemented simply in Arc/Info and is routinelyused in ArcView and other GIS packages. Each spatiallayer also includes digital registration tics.

Associated with each of these feature types are threedatabase tables containing attributes. The first is the fea-ture attribute table, which contains one record for eachspatial feature (such as a fault). This table contains identi-fication numbers for each feature, topology fields (if any),area, perimeter and(or) length as appropriate, a unique pri-mary key field (ITEMID), and a single characteristicdescriptive attribute (see below). The feature attributetable for polygons is termed a Polygon Attribute Table(PAT), for arcs an Arc Attribute Table (AAT), and forpoints a Point Attribute Table (PAT).

The second table is a definition table, also known as alookup table or domain table. This table contains expand-ed definitions for each unique attribute in the correspond-ing feature attribute table. Therefore the relationship ismany-to-one, the “many” being the set of records in thefeature attribute table that share a common attribute (e.g.

all polygons with ULABEL of Kgr), and the “one” beingthe single record in the definition table that describes thatattribute (e.g., Cretaceous granite of Hidden Wells). Eachdefinition table can contain additional items, including car-tographic symbol assignments.

The third table is an ID Table, which can contain addi-tional attributes, such as strike and dip for a structuralmeasurement, for each individual spatial feature. The rela-tionship is one-to-one with the corresponding featureattribute table. These attributes are placed in the ID Tablerather than in the feature attribute table to keep that tableas small as possible. This enhances the speed of manyGIS operations, and makes the equivalent feature attributetables of every map database identical in structure, whichpermits easy mosaicking. Any additional attributes areplaced in the ID Table, whose structure can vary betweenadjacent maps as required.

The units, contacts and faults (UCF or geology) layeris required for all geologic maps, and contains arcs andpolygons. Polygon and arc topology must be present.Arcs represent contacts and faults (as well as water bound-aries and the map boundary); arcs that are not part of apolygon boundary (dangling arcs, e.g. some faults) arepermitted. Arcs with polarity (e.g. low-angle faults) mustbe encoded with right-hand rule. The characteristicattribute in the UCF.AAT is LTYPE (for line-type, e.g.thrust fault or contact); the definition table, UCF.LN, adds

DATA MODEL FOR SINGLE GEOLOGIC MAPS: THE NATIONAL GEOLOGIC MAP DATA MODEL

Structural Data (SD) Layer

SD.PAT SD.AAT

ITEMID *

STYPE

System-specific attributes

SD.STSTYPE *

SDEF

Other attributes...

ITEMID *

SD.PID

Attributes as needed...


ITEMID *

LTYPE

LTYPE *

LDEF

Other attributes...

SD.LN

ITEMID *

SD.LID


SMODE

AZIMUTH

INCLIN

Optional layerNo topology

PointNode

1

1

1

1

1 1

1-M 1-M

1

11

1

* = Primary key

SMODE*SMDEF

1-M

1

SD.SMDEF

Units, Contacts and Faults (UCF or geology) Layer

Required layerPolygon topology

UCF.PAT UCF.AAT

ITEMID *

ULABEL


UCF.UNULABEL *

UNAME

Other attributes...

ITEMID *

UCF.PID



ITEMID *

LTYPE

LTYPE *

LDEF

Other attributes...

UCF.LN

ITEMID *

UCF.LID


1

1

1-M

1

1

1

1

1

1

1-M

1

1

Polygon labelNode

Featureattributetables

Definitiontables

ID tables

Figure 1. Single Geologic Map (SGM) Data Model.

79

the LDEF, the line-type definition. Polygons representgeologic units and consist of bounding arcs and a singlelabel point internal to each closed polygon. The character-istic attribute in the UCF.PAT is ULABEL (for unit-label,e.g. Kgr, Qal). The definition table, UCF.UN addsUNAME, the formal or informal name of the map unit.

The structural data layer (SD) is optional and containspoints and lines; no topology is present (no intersectionswhere structural lines cross). Points represent structuralmeasurements such as attitudes. The characteristicattribute in the SD.PAT is STYPE (for structure type, e.g.bedding or lineation). The definition table SD.ST addsSDEF, the full definition of each STYPE, and SMODE,the convention used to record orientations (e.g. strike anddip, trend and plunge). A separate table, the SD.SMDEF(for structure mode definition), defines each of thesemodes or conventions. The point ID Table (SD.PID) con-tains the structural measurements, AZIMUTH and INCLIN(inclination), for each point. Arcs in the structural datalayer represent structural lines such as fold axes. Thecharacteristic attribute in the SD.AAT is LTYPE (e.g. anti-cline). The SD.LN definition table adds LDEF, the defini-tion of each unique LTYPE.

The relationship of the SGM data model to theNational Geologic Map data model is straightforward.The spatial layers and their associated feature attributetables correspond to the Spatial Object Archive in theNGM data model. The SGM Definition Table is a simpli-fied subset of the tables in the NGM Compound ObjectArchive. The SGM ID Tables correspond to the tables inthe NGM Singular Object Archive, with the difference thatthe SGM data model currently allows a single record inthe ID Table for each spatial feature. In a later SGM ver-sion, we expect to provide for multiple records referring tosingle map elements in order to accommodate coincidentfeatures such as multiple observations at a locality.

CONCLUSIONS

The SGM data model has been developed specificallyto address the practical problems encountered on a dailybasis as we work to develop and deliver digital geologicmaps. It is designed to serve our immediate needs for acommon storage and exchange format and to be expandedas we develop consensus on how best to encode otherparts of a geologic map. We expect this expansion tomove the data model progressively toward the full NGMdata model. The SGM model must be immediately practi-cal, and as a first step in implementation we have preparedAML tools for use in Arc/Info to convert between the ALCand SGM models and to operate on the SGM definitiontables in ALACARTE. Further tools, converters, and con-tinuing support for the SGM data model will be developedand provided by the team as needed, with the ultimate goal

of having tools with which to compile, edit, query, andpublish directly in the SGM model.

We view the SGM model as a means to present a stan-dard face to each other and our users, but also as a mecha-nism by which to draw our GIS practitioners together andto stimulate the joint development of tools and proceduresof common value. We invite participation by others in thisdevelopment process and welcome active participation indevising, testing, and adopting expansions to the modeland to the tools and procedures with which to use it. Thedescriptions, code, proposed expansions, and discussionsare available at our GIS web sitehttp://geology.wr.usgs.gov/techniques.

ACKNOWLEDGMENTS

I write from my vantage as team leader, but ourprogress toward digital consensus, and much of the detailherein, have depended on the constructive work of numer-ous participants, including David Bedford, Debra Block,Todd Fitzgibbon, Scott Graham, Russell Graymer, RalphHaugerud, Steven Kennedy, Jon Matti, David Miller,Geoffrey Phelps, Carl Wentworth, and Karen Wheeler.

REFERENCES

Brabb, E.E., Graymer, R.W., and Jones, D.L., 1998, Geology ofthe Palo Alto 30 X 60 minute quadrangle, California: a digi-tal database: U.S. Geological Survey Open-File Report 98-348, <http://wrgis.wr.usgs.gov/open-file/of98-348>.

Fitzgibbon, T.T., 1991, ALACARTE installation and system man-ual (version 1.0): U.S. Geological Survey, Open-FileReport 91-587B,<http://wrgis.wr.usgs.gov/docs/software/software.html>.

Fitzgibbon, T.T., and Wentworth, C.M., 1991, ALACARTE userinterface - AML code and demonstration maps (version1.0): U.S. Geological Survey, Open-File Report 91-587A,<http://wrgis.wr.usgs.gov/docs/software/software.html>.

Matti, J.C., Miller, F.K., Powell, R.E., Kennedy, S.A.,Bunyapanasarn, T.P., Koukladas, Catherine, Hauser, R.M.,and Cossette, P.M., 1997a, Geologic-point attributes for dig-ital geologic-map data bases produced by the SouthernCalifornia Areal Mapping Project (SCAMP) Version 1.0:U.S. Geological Survey, Open-File Report 97-859.

Matti, J.C., Miller, F.K., Powell, R.E., Kennedy, S.A., andCossette, P.M., 1997b, Geologic-polygon attributes for digi-tal geologic-map data bases produced by the SouthernCalifornia Areal Mapping Project (SCAMP) Version 1.0:U.S. Geological Survey, Open-File Report 97-860.

Matti, J.C., Powell, R.E., Miller, F.K., Kennedy, S.A., Ruppert,K.R., Morton, G.L., and Cossette, P.M., 1997c, Geologic-line attributes for digital geologic-map data bases producedby the Southern California Areal Mapping Project(SCAMP) Version 1.0: U.S. Geological Survey, Open-FileReport 97-861.


McKee, E.H, Wickham, T.A., and Wheeler, K.L., 1998,Evaluation of faults and their effect on ground-water flowsouthwest of Frenchman Flat, Nye and Clark Counties,Nevada: a digital database: U.S. Geological Survey Open-File Report 98-580, <http://wrgis.wr.usgs.gov/open-file/of98-580>.

Miller, F.K., Matti, J.C., Brown, H.J., and Powell, R.E., 1998,Digital geologic map of the Fawnskin 7.5’ quadrangle, SanBernardino County, California: U.S. Geological SurveyOpen-File Report 98-579, <http://wrgis.wr.usgs.gov/open-file/of98-579>.

Johnson, Bruce R., Broderic, Boyan and Raines, Gary, 1998a,Digital Geologic Map Data Model (version 4.2):

Unpublished American Association of State Geologists /U.S. Geological Survey draft document,<http://ncgmp.usgs.gov/ngmdbproject/>.

Johnson, Bruce R., Broderic, Boyan, Raines, Gary, Hastings,Jordan T., and Wahl, Ron, 1998b, Digital Geologic MapData Model Addendum to Chapter 2 (version 4.3):Unpublished American Association of State Geologists /U.S. Geological Survey draft document,<http://ncgmp.usgs.gov/ngmdbproject/>

Wentworth, C.M., and Fitzgibbon, T.T., 1991, ALACARTE usermanual (version 1.0): U.S. Geological Survey, Open-FileReport 91-587B,<http://wrgis.wr.usgs.gov/docs/software/software.html>.

DATA MODEL FOR SINGLE GEOLOGIC MAPS: THE NATIONAL GEOLOGIC MAP DATA MODEL 81

ABSTRACT

We have developed an internet-based digital libraryprototype (CordLink) to seamlessly integrate digital maps,images and text on Canadian Cordilleran geology into acomprehensive information resource for the geoscientistand non-geoscientist. The library structure addresses sev-eral factors that make geological knowledge challenging touse in an internet environment: (1) geological data is mul-tidisciplinary and diverse (e.g. bedrock mapping, geo-physics, geochronology, mineral resources, etc.); (2) it isexpressed in multiple forms such as map, text, and image;(3) it is extremely inter-related with no piece of knowledgecaptured in any one fragment or form; (4) it spans geo-graphic space and geologic time; (5) it is expanding andevolving; and (6) users tend to peruse library holdingswithin a specific geographic, geologic, or level of exper-tise context. These factors require an approach to datamanagement that, while it is distributed in nature, is morestructured and context-sensitive than the simple linking ofa myriad of web pages.

The CordLink holdings are held in a relational data-base system arranged according to an extended version ofthe proposed US National Digital Geologic Map DataModel design (<http://ncgmp.usgs.gov/ngmdbproject>).The database is coupled to the Internet-based AutodeskMapGuide Geographic Information System (GIS). Userscan enter the holdings from the Map, Document, Image,

Education or Research perspectives and can then navigatethe library according to subject while preserving their ini-tial focus. Derivative thematic maps and reports can beconstructed and generated by the user on-the-fly for supe-rior browsing capability. Various library holdings can bedownloaded and users can contribute elements to thelibrary. A catalog of on-going research projects is provid-ed as well as an on-line scientific discussion forum. It isanticipated that this wide-ranging, tightly inter-linkedresource, delivered in a visual environment, will benefitresearchers, students, instructors and the general public.

INTRODUCTION

The Internet possesses enormous potential to enhancethe distribution of geological information, and to broadenthe scope of its usage. The convenience of the Internetensures that geological information will be accessed moreoften and more diversely than ever before, and it will nodoubt attract more non-traditional visitors to geologicalinformation than, for example, did the libraries or salesoffices of geological surveys. In this, however, it couldserve to alienate as many people as it attracts, for, likemany professions, geology is a field that is difficult tocomprehend by the uninitiated. This is disturbing in lightof the attention that geological agencies are now directingto outreach and societal benefit. Note, for instance, that

83

Using the Proposed U.S. National Digital Geologic Map DataModel as the Basis for a Web-Based

Geoscience Library Prototype

By Boyan Brodaric1, J.M. Journeay2, and S.K. Talwar2

1Geological Survey of Canada615 Booth Street

Ottawa, Ontario K1A 0E9Telephone: (613) 992-3562


2Geological Survey of Canada101-605 Robson Street

Vancouver, B.C. V6B 5J3

the strategic directions of both the Canadian (GSC) andUS (USGS) geological surveys emphasize this societaltheme:

“Its (GSC’s) goals are threefold:• to create centres of excellence in Canada for scien-

tific research and the advancement and dissemi-nation of knowledge,

• to better apply our scientific knowledge in achiev-ing Canada’s economic and social goals, and

• to contribute to a better quality of life forCanadians.” (GSC, 1996).

“The challenge for the USGS is to stay focused ona horizon of some ten years out, while realizing thatthere will be near-term shifts… Beyond thesealready compelling factors are the public’s percep-tion of its investment in science as a means of solv-ing societal problems and society’s concept of the“public good” of science.” (USGS, 1997).

Therefore, the challenge of the Internet extendsbeyond making geological information available, for thatis readily done; it must instead make this knowledge moreuseful to geologists and more relevant to other parts ofsociety.

It seems clear, at least to many geologists, that funda-mental geological investigation can play an important rolein understanding and managing the relationship betweenhumans and the natural environment. This role, however,is not readily evident in traditional geological productssuch as maps or databases, but must be teased out of themby geological experts. This effectively undermines thebroader use of the information and diminishes its value tothe non-geologist. Combining modern GIS tools with theInternet yields technology capable of overcoming this situ-ation; however, how to structure, present and manipulategeological knowledge, in its fullest sense, to other parts ofsociety, is not clear.

What does seem clear is this: if the Internet is to serveas the gateway to the public, it is incumbent on geologicaldata providers to ensure that geological knowledge is rep-resented to its fullest degree, and that it is accessible andusable by professional geologists and casual visitors. Thisrequires two major interacting components, one represen-tational and the other functional: firstly, a data model forgeological knowledge is required to meaningfully structuregeological information, and secondly, a suite of operationstuned to both geologists and non-geologists is necessary toleverage such a structure. The remainder of this paperdescribes these two elements. The requirements for acomprehensive geologic site are discussed first, followedby the data representation methods required to make itwork. Finally, an Internet site that endeavors to meet thesecriteria, by serving the geology of Canadian Cordillera, isdescribed.

REQUIREMENTS

Data, Information and Knowledge

What we know is often categorized as data, informa-tion and knowledge. In a geological sense we can see thisas an increasing gradient in our understanding of the inter-action of earth processes and materials: data are measure-ments, facts and observations; information is data in con-text, such as the interpretations on a geologic map; andknowledge represents the complete understanding that wepossess in our minds of some thing, such as a geologicalhistory of an area. Thus the computer representation ofreal knowledge is very difficult, and arguably impossible,as we can’t capture human thought (at least not yet). Themost we can do is mimic human thought via knowledgerepresentation strategies and reasoning mechanisms thatcreate the impression of real thinking—some would arguethat this indeed constitutes machine intelligence, if wecan’t distinguish between machine and human behavior(Turing, 1950). In terms of geological computing, thisimplies two things: firstly, that we must maximize ourknowledge representation efforts, and secondly, that wemust provide superior functionality so that the geologicalcontent that is represented is used intelligibly and willappear to be, in fact, knowledge.

Knowledge Representation

Knowledge representation for computing purposesrequires the objects and their relations in the real world tobe translated into objects and relations in the computingworld (Luger and Stubblefield, 1998, p. 293). In effectthis states that knowledge can be approximated in a com-puter by carefully cataloging and indexing information(and then using it). We must therefore carefully identifythe geologic objects to be represented and maximize therelationships between them. This is analogous to placingthe information in as many contexts as possible, and isparticularly relevant to geological mapping where informa-tion can be viewed quite differently according to a geolo-gist’s expertise, education or bias. In non-digital environ-ments the expression of a viewpoint is usually a multi-media affair, often requiring maps, reports, charts, dia-grams and oral presentations to convey a message. Howcan computing mechanisms cope with such diversity?

At a gross level, for computer representation purposes,we can also attempt to categorize the objects of geologicalknowledge to be maps, documents, images, projects, andreferences. Though it is worthwhile to view a map along-side its accompanying reports, diagrams, bibliography, andgeneral project description, it is much more useful to actu-ally interconnect the sub-components of these elements topermit investigation at a more detailed scale. For instance,selecting a polygon on a map and viewing its unit descrip-


tion, embedded in reports or articles, as well as displayingrelated images such as age correlation charts or cross-sec-tions, or obtaining references specific to that map unit,would be much more beneficial. Narrowing our scale offocus to field observations would be of even greater bene-fit, though the task would be much more difficult as therelationships between field observations and their interpre-tations is multitudinous and complex. This suggests thebasic resolution for the represented objects could be quitefine, and their relationships abundant and intricate.Designers of geological software must select an appropri-ate scale of representation based on their purposes, and onthe quantity and quality of the information content. Amedium scale of object resolution could include:

• Map: legend, map unit, occurrence.• Geological Description: lithostratigraphic age,

geochronologic age, lithology, and others.• Document: volume, chapter, subsection, figure.• Image: caption and image.• Project: originating project details.• Reference: citation and source details.

The ideal system would permit relationships betweenany of these elements to be represented. For instance, onemight relate a lithostratigraphic or geochronologic age to abody of text, to a figure, to a map unit or to a map.Likewise a map unit might be associated with one or moregeological descriptions, document segments, figures, pro-jects or references. The ability of establishing such rela-tionships elevates the context of any single informationfragment and thus increases its knowledge content.Furthermore, this design is scaleable as geological descrip-tions could be expanded to contain the more complex fieldobservations.

Knowledge Usage

A fitting model for how geologic objects are to beused, rather than what they are, is the library: libraries arevalued for their archival role as well as for their ability tostimulate new thought. Library contents possess thepotential to foster new insight and thus to regenerate them-selves with new contributions. This implies that informa-tion requires discourse to achieve the rank of knowledge.Applying this to digital environments suggests digitallibraries must not only represent original author intentionbut must promote scientific discourse by enabling the rep-resentation of different viewpoints. They should be inclu-sive of diversely formatted and conceived documents, andalso encourage various methods of discussion. Todaythese ideas are perhaps best exemplified by the Internet, itsinterconnection of computers known as the world wideweb, and its loose arrangement of information fragments.Many researchers suggest that this arrangement is not opti-mal (e.g. Schatz, 1995), and contend that the web requires

mechanisms to add context to its information fragments,which essentially means increasing the resolution of itsobjects and relations. Why is this so ?

Implementation

The world wide web is a form of knowledge manage-ment system at the document level. It primarily maintainslinks between documents (consisting of images and text),and permits searching for text strings within the docu-ments. A first order approach to placing geological infor-mation on the web would be reduce the geological infor-mation to a set of text and image documents that are inter-linked, and simply use the web’s navigation capabilities totraverse them. There are several problems with thisapproach:

1. Spatial representation: Maps are treated as diagramsrather than spatial constructs. Smart graphic formats(e.g. CGM) will allow individual elements to retainlinkages to other document components, but howadjoining maps fit together, for instance, is beyond thebasic web.

2. Spatial operation: searching for information inside amap unit or close to a fault is impossible.

3. Contextual searching is impossible: searching for theinclusion or exclusion of words within a document isthe limit; applying contextual parameters is impossi-ble: e.g. searching for map units of a certain age willreturn any geological or non-geological document thatcontains the specified text.

4. There is no provision for the management of informa-tion. For instance, no referential integrity is imple-mented—this means, for example, that the spelling ofterminology could not be standardized, which causessearches to return incomplete results.

Thus, we can view the web as a very valuable docu-ment server, but not as an analytic tool. Missing is thegeological context that is crucial to its effective use.Industry and academia are spending significant resourcesto address these problems, through various digital libraryinitiatives (see National Coordination Office forComputing, Information and Communications, 1998).Although considerable progress has been made, the con-textual web is not yet a reality today. This suggests thatpractical applications must, in the short term, implementtraditional technologies to manage the objects and rela-tions of interest to us, and this implies that database sys-tems and GIS must be harnessed to the Internet to serve asour knowledge engines. Of the two technologies, only thedatabase system permits conceptual extensions, as GIS arefixed in one spatial model or another. For example, vec-tor-based GIS typically offer points, lines and polygons,

USING THE PROPOSED U.S. NATIONAL DIGITAL GEOLOGIC MAP DATA MODEL 85

and the topological relations between, as their spatial prim-itives. The user-defined meaning of these spatial objectsand their relations is typically contained within a database.

From this it is apparent that it is increasingly impor-tant to develop and adopt a database design that containsrelevant geological objects and encourages effective usageof them. The medium scale objects noted above (maps,descriptions, documents, images, projects and references)provide a good starting point for a database design that isknowledge-oriented. How well do our current data model-ing efforts meet these criteria?

GEOLOGICAL MAP DATA MODEL

The geological map has evolved into a highly concisearticulation of the geological history of a specific geo-graphic region. Though its graphic nature makes it emi-nently suitable for web-based knowledge delivery, its com-plexity and diversity challenge simplistic conceptions ofwhat a digital geologic map is, and how it functions. TheU.S. National Digital Geologic Map Data Model effortaddresses the first of these issues, by defining the compo-nents of a geologic map. The prototype data model 4.3(<http://ncgmp.usgs.gov/ngmdbproject/>) is designed torepresent the most common elements of geologic mapdata, and is thus well suited to representing Map,Geological Description and Reference objects and certainrelations amongst them:

1. Reference sources, including maps, as well as legends,legend items (e.g. rock units) and their relations, aredefined thematically and cartographically. This per-mits geological map information to be organizedcoherently, in databases, while preserving their maporigins.

2. Geological descriptions consisting of lithologies andages (geochronological and lithostratigraphic) areassociated with legend items or individual map occur-rences (e.g. with one rock unit polygon, or one thrustfault line).

3. Geologic vocabularies for rock types, time scales andstructural features can be defined, extended and stan-dardized. The use of hierarchies enables the develop-ment of terminological standards that require consen-sus at higher levels but that may be modified accord-ing to user need at lower levels.

4. Geologic information is separated from spatial repre-sentation, permitting a geological feature to have dif-ferent spatial representations (point, line, polygon,volume, etc.) by existing on different maps, at variousscales, or because of diverse interpretations.

However, in keeping with the stated intention of stor-ing knowledge by maximizing geological context throughenhanced representation (objects and relations) and func-tion (digital library operations), several aspects of thisdesign require modification:

1. Objects of representation—missing are document,image, project and dataset objects:

A. Document: a repository for bodies of text comprising reports, journal articles, theses, etc.

B. Image: a repository of images including georeferenced remote-sensed maps (e.g. geophysics), as well as diagrams from a map’s surround, or figures from a body of text.

C. Project: a record of recent, current and future geological activities, their purposes, goals and locations.

D. Dataset: a description of the physical layers comprising a map including supplemental informationabout a dataset such as its name, location, layer type (e.g. line, point, polygon).

2. Relations between objects:

A. Description independence: this requires establishing a description archive where geological descriptions, such age ranges, rock compositions, and other description types (e.g., images, text, projects, mines, wells, boreholes, etc), exist as independent entities that may optionally be related to one or more rock units or to other descriptions. In v4.3 this is not the case, as age ranges and rock compositions must be related only to a specific rock unit.

B. Description relations: denotes the complex relationships between geological descriptions. It would, in effect, permit any description to be related to any other description. This is of specific importance to medium scale geologic knowledge representation as it permits, for example, text fragments to be associated with ages, images,projects, field sites, measurements, etc., thereby cross-indexing the geological content according

to different perspectives.

C. Subject indexing: a generic, hierarchical, subject listing that can be used to index digital library holdings and thus provide novice as well as expert search capability.

Modifications 2a and 2b have been discussed withinthe former U.S. national geologic map data model working


group (<http://ncgmp.usgs.gov/ngmdbproject/>) and havebeen adopted by the CordLink library. The remainingmodifications result from implementing a digital libraryapproach to geologic information within the CanadianCordillera (Brodaric and others, 1999).

A DIGITAL LIBRARY IMPLEMENTATIONFOR THE CANADIAN CORDILLERA

A large database of geologic maps for the CanadianCordillera was constructed according to a data modeldesign (Figure 1) modified from the proposed U.S. nation-al v4.3 prototype. The database represents a digital librarythat contains five main geologic map series ranging inscale from 1:2,000,000 to 1:32,000,000 and covering alarge geographic area including all of British Columbia,the Yukon, and parts of Alberta. These maps contain inexcess of 210,000 individual map features, as well asabout 3500 legend items, mostly rock units. Each rockunit is described according to absolute age range (e.g. 300-400 million years), geologic age range (Devonian-Siliurian), rock type composition (shale, siltstone, etc.),and is also linked to a cartographic symbol within a leg-end. The Cordilleran volume from the definitive Decadeof Geology of North America (DNAG) (Gabrielse andYorath, 1992) was converted to digital format and com-partmentalized, including more than 800 pages of text and500 images. Text components were indexed according to

their geologic age and subject, and to relevant figures andimages, and to related rock. Users are thus able to retrieveauthoritative descriptions, including text and images, whenviewing any feature instance on a map; users are also ableto enter the archive from the text and image perspectives,with all interrelationships maintained. The description andlocation of research projects can be viewed on-line, andusers can enter their own research project information. Adiscussion room is being provided to encourage discourseon geoscience topics of current interest. Some map datamay be downloaded and an education component main-tains links to education sites within the geosciences.

The library is constructed to operate in an Internetenvironment, by utilizing the Autodesk MapGuide soft-ware (<http://www.mapguide.com>) to display and containits spatial aspects; the remaining non-spatial aspects of thedatabase are hosted within the MS SQL Server relationaldatabase environment. These two components were inte-grated into an attractive and functional user interfacethrough extensive JAVA programming, performed by anexternal contractor.

The library can be viewed using MS Internet Explorerv4.0 or Netscape 4.5 once the appropriate plug-in isinstalled. Although the web site’s definitive address is stillunknown, the CordLink digital library is accessible via<http://www.rgsc.nrcan.gc.ca> (follow the CordLink link).It is anticipated the site will enhance geological researchand geological education through the digital interconnec-tion of the various map, text and image components.


Figure 1. A diagram depicting the enhanced digital geologic data model utilized by the CordLink project

The web site is arranged according to 8 main mod-ules: Maps, Documents, Images, Research, References,Data Download, Education, Search.

Home

The Home page (Figure 2) provides an entry point tothe main modules, as well as an overview of the siteincluding information about its intent, content, systemrequirements, its development team and other partners.

Maps

The Maps module allows maps and ancillary data tobe viewed and queried at various scales. The geologicmap and ancillary data available is tailored to the viewingscale: regional views will depict general maps and data,and as the user zooms into a detailed area more relevant,detailed information is provided. Users are able to perusethe holdings according to a subject index or geospatially,through a scale-sensitive legend. Table 1 provides a par-tial, representative, listing of the available maps anddatasets.

The Maps module also enables map features to be the-matically filtered and inspected according to their legenddescription (e.g. rock unit, age or lithology) and relateddocument fragments and images. For example figure 3adepicts a portion of the 1:2,000,000 tectonic assemblagemap of the Canadian Cordillera (Journeay and Williams,1995) where a set of map features containing Basalt, atleast in part, were identified (outlined with dashed lines).

Figure 3b illustrates how a specific polygon in this set isdescribed in terms of rock unit (i.e. Cache Creek), DNAG(Gabrielse and Yorath, 1992) text and image.

Documents, Images and Research

The Documents, Images and Research modules areanalogous to the Maps module: their contents are indexedaccording to subject, and provision is made for selectingand viewing module contents according to related map,image, text or project information. Of course, the primaryviewing window is focused on the module’s media type: amap is displayed in the Maps module, text is displayed inthe Documents module, an image is displayed in theImages module, and a project location map is displayed inthe Research module. Smaller windows then present theremaining media type information. The Images andResearch modules additionally permit users to insert theirown contributions to the library via an on-line interface,whose content is reviewed by a digital librarian.Currently, image contributions must be transferred to theCordlink site, though it is expected they could remainremotely located in the future. The research module alsomaintains an on-line discussion platform that encouragesscientific discourse.

References, Data Download,Education, and Search

The References, Data Download, Education andSearch modules provide several important functions:


Figure 2. The Home page for CordLink—a digital library prototype for the Canadian Cordillera.

• References: provides access to references within theCordLink site, and links to other Canadian geosciencereference libraries.

• Data Download: currently permits a select subset of theavailable maps to be copied by the user to their localcomputer, in a variety of common geospatial formats.

• Education: provides access to on-line educationresources related to Canadian goescience.

• Search: a generic mechanism to locate document levelinformation (i.e. metadata) within the library.Document level information refers to title, author, sub-ject, etc., descriptions of maps, reports, images andother documents; it does not refer to their internalcontents.

FUTURE DIRECTIONS

The CordLink project is one component of theResSources GSC digital geoscience initiative within theGeological Survey of Canada(<http://www.rgsc.nrcan.gc.ca>). The goal of this initia-tive is to link significant GSC data holdings across theInternet, in order to facilitate their dissemination, integra-tion and general interoperability. The ReSSources GSC

initiative is itself part of a larger Canadian geoscience pro-gram, the Canadian Geoscience Knowledge Network(CGKN), which has similar goals and includes all themajor Canadian federal, provincial and other geosciencedata providers. CGKN is in turn a part of the CanadianGeospatial Data Infrastructure (CGDI), a national effort tomake all geospatial data accessible and useable on theInternet. As CordLink is contributing significant regionalCordilleran geology holdings to these efforts, it is expectedthat the site will continue to evolve. In particular, it isexpected that its current prototype status will be upgradedto a fully operational site in the next year. Longer termconcerns include:

• expanding the distributed nature of the library, and

• enhancing its interoperability with other libraries,

without sacrificing the gains made in representing andusing greater geological context. Advances in digitallibrary research will resolve some of these issues and theCordLink library should be well positioned to adapt thesolutions to its holdings and infrastructure.

CONCLUSIONS

A digital library approach to representing and deliver-ing geological knowledge was implemented for Canadian


Author Date Title

H.Gabrielse and C.J. Yorath 12/31/91 The Geology of the Cordilleran Orogen in Canada

Wheeler J.O., Hoffman, Card, K., Davidson, T., 3/31/97 Geological Map of Canada - 5MSanford, Okulitch, and RoestKirkham, R.V., Chorlton L.B. and Carriere J.J. 1/30/95 Generalized Geology of the World – 32MFulton, R.J. 1/30/97 Surficial Materials of Canada - 5MJourneay J.M.and Williams S.P. 3/30/95 GIS Map Library: A Window on

Cordilleran Geology – Tectonic Assemblages 2M

Journeay J.M. and Williams S.P. 3/30/95 GIS Map Library: A Window on Cordilleran Geology – Terranes 2M

Journeay J.M. and Monger, J.W.H. 3/30/98 Coast Belt Geoscience Library – 250KB.C. Geological Survey Map holdings, Mineral Inventory,

Assessment ReportsB.C. Ministry of Environment 12/31/94 TRIM topographic base dataVarious 12/31/94 topographic dataVarious 4/1/99 Shaded Relief – DEMVarious 4/1/99 Satellite ImageVarious 4/1/99 Bouger GravityVarious 4/1/99 AeromagneticVarious 4/1/99 Earthquake epicentresC.J. Hickson and B. Edwards VolcanoesG.J. Woodsworth Geothermal resources

Table 1. A partial, representative, listing of geoscience data found in the CordLink digital library.


Figure 3a. The Map module displaying 1:2M tectonic units comprised, at least in part, of Basalt; the pointerhas selected one specific polygon for further interrogation.

Figure 3b. The library contents of the polygon selected in Figure 3a are displayed as database table, DNAGtext and figure.

Cordilleran geology. The digital library utilized the pro-posed U.S. national geological data model to augment thegeological context of the data holdings in terms of data-base representation and end-user functionality. The pro-posed U.S. national digital geologic map data model wasexpanded to accommodate more geological data types andricher relations between the new, and previously defined,data model components. A prototype Internet-based inter-face was developed to exploit the geological data typesand their rich relations. The results have been very satis-factory, in terms of proof of concept, and in terms of pro-totype implementation, and we anticipate developing thesite further. It is hoped the digital library approach willpromote the use of geological information, and knowledge,by using technology to enhance scientific investigation andthus to better meet the broader societal demands facing ustoday.

REFERENCES

Brodaric, B, Journeay M, Talwar, S., and Boisvert, E, 1999,CordLink Digital Library Geologic Map Data ModelVersion 5.2, (<http://www.rgsc.nrcan.gc.ca>: CordLink I,About CordLink), 29 p.

Gabrielse, H., and Yorath, C.J., eds., 1992, Geology of theCordilleran Orogen in Canada, vol. G-2, Decade of North

American Geology: Canada Communications Group,Ottawa, Canada, 844 p.

Geological Survey of Canada, 1992, A Summary of theGeological Survey of Canada’s Strategic Plan forGeoscience 1996 – 2001,<http://www.NRCan.gc.ca:80/gsc/strplan_e.html>.

Journeay J.M. and Williams S.P., 1995, GIS Map Library: AWindow on Cordilleran Geology – Tectonic AssemblagesLGeological Survey of Canada Open File 2948, Ottawa, ON.

Luger, G.F., and Stubblefield, W.A., 1998, Artificial Intelligence;Structures and Strategies for Complex Problem Solving:Addison-Wesley, Reading, MA, 824 p.

National Coordination Office for Computing, Information andCommunications, 1998, Technologies for the 21st Century;The Digital Libraries Initiative,<http://www.ccic.gov/pubs/blue98/dig_libraries.html>.

Schatz, B.R., 1995, Information Analysis in the Net: TheInterspace of the Twenty- First Century. White Paper forAmerica in the Age of Information: A Forum on FederalInformation and Communications R & D, July 6-7, NationalLibrary of Medicine,<http://www.canis.uiuc.edu/papers/america21.html>.

Turing, A., 1950, Computing machinery and intelligence: Mind,59, 433-460.

U.S. Geological Survey, 1997, USGS Strategic Plan 1997-2005,<http://www.usgs.gov/strategic>.


93

INTRODUCTION

One objective of the Geologic Mapping Act (GMA) of1992 (re-authorized in 1997) calls for the establishment ofstandards for digital geologic mapping both for paper plot-ting and for a computer-readable database. One responseof the U.S. Geological Survey (USGS) and the NationalCooperative Geologic Mapping Program (NCGMP) toOffice of Management and Budget Circular A16,Executive Order 12096 and the GMA, will be to establisha national digital geologic map database for use at a scaleof 1:100,000. This work is to be conducted under theNational Geologic Map Database Project (NGMDB) of theNCGMP.

In August 1996, in St. Louis, Missouri, the DigitalGeologic Mapping Committee of the Association ofAmerican State Geologists (AASG) and NGMDB of theUSGS, formed several working groups to devise standardsand guidelines for various concepts that make up a geolog-ic map in digital form. Information about these workinggroups is available at the NGMDB project web site,<http://ncgmp.usgs.gov/ngmdbproject>. One element ofthe geologic map database standards effort that has had lit-tle attention is map geometry.

The NGMDB needs to test the concept of a nationalgeologic map database at a scale of 1:100,000. TheGreater Yellowstone Area (GYA) is a region in which sucha geologic database, built as a proof of concept, would addmuch to the understanding of the GYA ecosystem. TheGYA encompasses an area five degrees in longitude (108West to 113 West) and four degrees in latitude (42 Northto 46 North). Forty 30’ by 60’ 1:100,000-scale quadranglemaps cover the GYA (figure 1). I anticipate that this pilotdatabase will be used to supply geologic data for the GYAScience Initiative, a new part of the Integrated NaturalResources Science program (INATURES) of the USGS,

which has become a part of the Department of Interior’sPlace-Based Science Program.

The NGMDB has the opportunity to test the results ofits standards efforts on a database that is needed in theregion by various interest groups, and that when integratedwith other geospatial data sets of the GYA will prove to bevital for ecosystem management there. However, to insurethe continued usefulness of the database, the geologic datain the database must be updatable and the GIS systemmust be able to keep careful track of the revisions as theyare made. In addition, digital map databases from the stategeological surveys of Idaho, Montana, and Wyoming willbe integrated into the database. Each state survey willmost likely have different ways of representing the geome-try of their products.

To support the GYA database effort, I am convertingunpublished “legacy” geologic map materials of W.G.Pierce for the geologic map of the Cody, Wyoming 2-degree sheet that lies on the east side of YellowstoneNational Park in the northeast corner of the GYA (figure1). These materials were originally compiled at a scale of1:125,000. The density of line work and the level of geo-logic detail allow the capture of the data at a scale of1:100,000. In particular, I am working with the unpub-lished Cody, Wyoming 1:100,000-scale quadrangle, whichis in the northwestern quadrant of the Cody 2-degree map.

The Cody, WY 1:100,000-scale geologic map dataexists as a film positive that was compiled by W.G. Pierce(1997) for the geologic map of the Cody 2-degree quad-rangle (figure 2). Time constraints preclude the use of theseven published 15’ geologic quadrangle maps as the pri-mary source materials for the 1:100,000-scale map. Thethree other 1:100,000-scale pieces of the Cody 2-degreesheet are also unpublished and exist only as film positives.Large-scale source materials for these maps have yet to belocated.

The Geometry of a Geologic Map Database

By Ronald R. Wahl

U.S. Geological SurveyBox 25046

Denver Federal Center, MS 913Denver, CO 80225

Telephone: (303) 236-1320Fax: (303) 236-0214


GEOMETRY AND THE DIGITALGEOLOGIC MAP

Problems concerning the capture of the geometry of ageologic map database have arisen as I continue to com-pile the Cody 1:100,000-scale geologic map. The prob-lems fall into three general categories. They are:

1. Capture of the “art” and at times subtle geologic con-cepts that the original compiler wished to portray.

2. Map-capture resolution and positional accuracy, and therelationship to National Map Accuracy Standards(USBB, 1947) and National Map Revision Standards(USGS-NMD, 1998).

3. Structure of the digital geologic map data and thespecifics of a data exchange or transfer format toallow users (within and outside the GYA) with a num-ber of different viewers and GIS tools to utilize thedata easily.

It is clear, that although geospatial data can be trans-ferred using standard transfer formats, the representationof geospatial objects such as polygons and lines containedin a digital geologic map database is not standardized.

CURRENT STATUS OF ACCURACYSTANDARDS

As of this writing no report has been released by any-one including the AASG/USGS standards groups men-tioned above, proposing any standards for map geometry

including standards for geologic map accuracy, or map res-olution. General policies for digital geologic map productshave been issued inside the USGS, but outside the USGSonly proposed guidelines exist for groups producing digitalgeologic map data (see Soller, Duncan, Ellis, Giglierano,and Hess, this volume). These guidelines do not mentionmap geometry. The OpenGIS consortium (OpenGIS,1999), in their abstract series of documents, has publisheda framework about which map geometry and accuracystandards for the digital representation of geologic mapsmay be formed. In addition, the National MappingDivision (NMD) of the USGS published a documentwhich details standards for updating already published1:24,000-scale topographic maps (USGS-NMD, 1998;Lemen, this volume). For now, the National MapAccuracy Standard (USBB, 1947) and the update docu-ment might be used to determine the accuracy with whichgeologic objects can be placed on a map until a moreappropriate standard is developed.

Problem 1: The capture of the geologic map“art” and concepts

Since the aforementioned 1996 meeting, I haveexpressed a concern that the “art” of a geologic mapshould be preserved as much as possible when the map isconverted to a GIS database. The following exampleshows one of the connections between geologic “art” andconcepts. Some 1:62,500-scale published geologic datanear Pat O’Hara Mountain in the southeastern part of theCody, WY 1:100,000-scale quadrangle, shows a number ofolder rocks units concealed by surficial materials. Whenthese data were compiled on the 1:100,000-scale map,some but not all of these surficial deposits were omitted(figure 3). I think that Pierce eliminated these surficialunits because he wanted to show the relationships amongthe older rocks and associated structures clearly. Maybe afuture compiler of the Cody 1:100,000-scale map mightwant to show these surficial units. In doing so, however,the compiler would again have to answer the question:“What are the important geologic concepts that I am tryingto portray and how might this best be done?”. My objec-tive in converting Pierce’s original map to a digital form isto faithfully reproduce what Pierce recorded on his origi-nal map.

In studying the data from this locale on the Codyquadrangle, I conclude that there are no standard ways torepresent linear data with characteristics more subtle thanthe primary attributes of “contact” or “fault”. An implicitassumption is made for rock units older (or different) fromQuaternary alluvium, that lines attributed as contacts showthe top of one formation and the bottom of another. Thisinformation is indirectly found in age attributes in the datamodel. But what happens if the ages that are given fortwo formations are both “lower Cretaceous” and they arein contact in one place but not in another on the same


-113

-113

-108

-108

46 46

42 42

Montana

Montana

Idaho

Idah

oWyoming

Wyo

min

gCody, WY

1:250,000-scalequadrangle

Cody, WY1:100,000-scale

quadrangle

Figure 1. The Greater Yellowstone Area.

THE GEOMETRY OF A GEOLOGIC MAP DATABASE 95

Fig

ure

2.

The

Cod

y, W

yom

ing

1:10

0,00

0-sc

ale

geol

ogic

map

.

map? I suggest that lines labeled as contacts be labeled inaddition with other information that would add geologicand geometric attributes such as “unit top” or “unit bot-tom”.

Another related question has arisen while capturingthe Cody geology. The lines were automatically generatedfrom a cleaned-up raster scan file by LT4X, a raster-to-vector conversion package used currently by the USGS-NMD. To reduce the angularity of such lines, the finalvector data set was smoothed using B-splines. The resultwas that, after the import into Arc/Info, the lines represent-ing contacts, faults, or dikes that had shown the slightest“wiggle” in them as raster lines had too many vertices.However, at map scale, the lines look “good” or smooth.

I am experimenting with the “GENERALIZE” and“SPLINE” commands in Arc/Info. I use the “GENERAL-IZE” command to eliminate the extra vertices in faults anddikes; and the “SPLINE” command with an appropriatevalue for the grain tolerance along with the “GENERAL-IZE” command with the “BENDSIMPLIFY” option toeliminate extraneous vertices in contacts. I am also exper-imenting with values for the “line simplification distance”option and the grain tolerance. Careful use of the “GEN-ERALIZE” command and the “SPLINE” command shouldproduce smooth contact lines and straight faults and dikesthat matches the original linework of the author.

Lastly, in some places, rock units that were mappedseparately on the 15’ geologic source maps are combinedwhen the outcrops of individual units are too small to beseen at 1:100,000. I have left them as Pierce compiledthem. I have done this so that the “art”, which is only par-tially map esthetics, and his geologic concepts would bekept intact.

Problem 2: Map resolution anddigital geology

The use of an autovectorizer such as LT4X to convertraster scans of map data to a vector format can introduceextraneous lines as vectors. One problem occurs at lineintersections. Line intersections are not always “clean” asa raster scan. As a result, Very Short Lines (VSL), linesless than 50 meters in length (at a scale of 1:100,000) are

inserted at some intersections, especially if two lines aspixels meet at a shallow angle (figure 4a). However, lineintersections are not the only case where VSLs are found.They also occur in places where line intersections as inter-preted from the scan are close together. When drawn on amap and then scanned, lines have an actual width that linesas vectors do not. In order to check for all VSLs, each linethat was 50 meters or less in length was selected and dis-played on a 1:100,000-scale plot of the line data. I thenhad to choose to delete the VSL or leave it in no matterwhat its origin was and perhaps even add to its length sothat a later check would not select this line (figure 4b).This latter choice might allow a fault to traverse an areavery close to other line intersections without intersectingnearby contacts at nodes and thereby change a geologicinterpretation.

The choice of 50 meters for a VSL was in origin achoice based on experience. I have learned since that theNMAS (USBB, 1947) document specifies that 90% of allobjects shall be within 1/50th of an inch of the true posi-tion when placing objects on a map at a scale smaller than1:20,000. This corresponds to 50.8 meters at a scale of1:100,000. A point of confusion may arise because USGS-NMD uses similar terms in the Federal Geographic DataCommittee (FGDC, 1998) metadata standard as follows:

1. Resolution (abscissa and ordinate): means the resolu-tion of the digitizing or scanning device. Since NMDuses digitizers and scanners that have a resolution of0.001 inches, NMD reports the resolution for1:100,000-scale DLG data as 2.54 meters. This is notthe same as:

2. Positional accuracy: NMD uses NMAS (USBB, 1947).

In addition, the Geography Department at theUniversity of California at Santa Barbara (UCSB) teachesa concept called “map resolution” in their GIS classes.UCSB defines map resolution to be linear distance belowwhich objects cannot be accurately located relative toother objects on a map of a given scale. The rule of thumbtaught to UCSB students is as follows:

(Map scale denominator) / 2*1000 = map resolution,in meters.

Then, map resolution would be 50 meters for a1:100,000-scale map, and 12 meters for a 1:24,000-scalemap. More recently, NMD states in the map revision doc-ument (USGS-NMD, 1998) that locations of revisions topreexisting maps especially from aerial photography maybe in error by as much as 22.2 meters (forty feet) of itsactual position on a 1:24,000-scale map. If this is scaledfor a 1:100,000-scale map, this positional inaccuracymight be as large as 88.8 meters. Moreover, one mightargue that the digital geologic map layer is a revision or an


Figure 3. Linework near Pat O’Hara Mountain, southernpart of the Cody, WY 1:100,000-scale map.

addition to a preexisting map, the topographic map. Inthis case, the accuracy with which digital geology can belocated on 1:100,000-scale topographic base maps may beon the order of 100 meters.

The understanding or lack thereof about these forego-ing concepts in this section by the geologic mapping com-munity has great consequences for digital geologic mapcompilation and its subsequent use as a thematic maplayer.

Problem 3: The transfer of digitalgeologic geometry

The compilation of lines and polygons representingthe geology of the Cody, WY 1:100,000-scale map will bedone in one layer or coverage. There are advantages tothis method of compilation. Lines that have multipleattributes both as linear features and polygon boundariescan be easily edited and updated. Unfortunately, this is notthe case for point data. While my study of the features onthe Cody materials shows no geologic features to symbol-ize as points, other 1:100,000-scale geologic maps includesuch features.

Mappers that I know usually delete rock outcrops thatshow as a point unless such features are a topographicallyor geologically important feature. They then “cartoon in”a line or polygon to represent these features. However,Reynolds (1971) used small triangles to indicate formationoutcrops that were too small to show on a 1:24,000-scalemap. The locations of these outcrops were important datathat show the reason that inferred unit contacts were drawnas shown on the map. It is not possible with most GISsystems to use point data formats to represent spatial fea-tures in a layer with polygon or line data. For example,any attempt to include data points in an Arc/Info coveragethat contains polygon and line data results in problems ifone issues a “BUILD” or “CLEAN” command. All of thedata points then become label points, and the effect ofthese commands cannot be undone.

Digital geologic map analysts prefer polygon and linedata in the same coverage while working in Arc/Info.However, a “casual” user of ArcView who imports such amap coverage will be surprised to learn that strictly lineinformation may be left behind. In ArcView, line data andpolygon data cannot exist in the same file. Other GIS sys-tems do not allow attribution of polygon boundaries withmore that one attribute. A line that is a polygon boundarycannot also be a fault in these cases.

The preceding discussion becomes more complexwhen one looks at existing and proposed standards for thesimple geometric terms: polygon and line. The definitionsof these terms differ among software vendors and stan-dards organizations. The interested reader can look atsome of the various standards listed below:

1. OpenGIS: <http//www.opengis.org/techno/specs.htm>

2. SDTS: <http//mcmcweb.er.usgs.gov/sdts>

3. SAIF: <http//www.elp.gov.bc.ca/~srmb/saif32/>

A possible solution to these problems comes first fromasking the question: What type of spatial data geometricobjects do modern GIS systems handle best? In my opin-ion, these systems handle polygons, of whatever type, best.Then notice that USGS-NMD publishes point data in DLGfiles by including points as lines composed of two verticesthat lie on top of one another. These two ideas lead to afurther idea that would allow the representation of bothpoint and linear features as polygons.

To represent points as polygons, image a circle whoseradius is twice the grain tolerance (in Arc/Info terms) (fig-ure 5a). The circumference could be attributed as a“scratch” line, which means that the circumference of thecircle would not be drawn. Marker symbols then coulduse the label point at the center to symbolize an attributeof the point. Circles of the recommended size would havelittle effect on the measured area of a containing polygonand would not be seen unless the circumference wasdrawn. I would call such a structure a “dot”.

Extending the idea to two dimensions (a line), imag-ine a linear feature that is enclosed by a “flexible” rectan-gle whose width would be four times the grain tolerancewith the line representing the linear feature running downthe center of the rectangle (figure 5b). This would maketwo polygons, one on either side of the central line thatcould be labeled “right” and “left” to show the directionthat the line should be traversed. I would call this con-struct a “wire”. Problems would arise with the implemen-tation of such constructs, but they would allow point andlinear features to be carried in the same coverage in allGIS systems. A wire would allow a linear feature to beclipped and still keep the directional information intact.Linear feature decoration would be applied to the centralline without interruption. The perimeter of the wire couldbe a scratch boundary when necessary. Unfortunately the


NodeNode

NodeNode

One pixel = 8 meters

One pixel = 8 meters

VSL

a. LT4X extraneous VSL b. Geologic feature of VSLs

Figure 4. Very short lines shown over original rasterscan.

“Dynamic Segmentation” feature of Arc/Info that wouldperform most of the same functions is proprietary and doesnot directly allow an import into other GIS systems.

The wire construct would be of great value in the geo-logic map database of the GYA, because there are five dif-ferent forms that dikes are recorded on geologic maps here(I am including sills in this discussion). They are:

1. Dikes that have a width that can be shown in true sizeon the map.

2. Dikes that are too narrow to show the actual width, butuse a polygon form to keep the geometry of the dikeand its relation to other rock outcrops and structures.

3. Dikes as lines. A line may represent more than onedike.

4. Dikes that are too numerous and too small to showindividually, but the map compiler wants to show thedike pattern directions and concentration.

5. Dikes that are too numerous and too small to showindividually and even the pattern of the dikes cannotbe seen, and the map compiler wants to show thepresence of dikes.

The wire construct would allow the geometric rela-tions of linear features like dikes and sills of the first threetypes to be kept in the geometry. The fourth and fifth diketypes can at present best be shown as overlay polygons inanother coverage. The number of dikes on the Cody1:100,000-scale map alone discourages the recording ofindividual geometric relations in an attribute relationstable. Such relationships might be more easily recorded ina form that is useable in a digital manner as a part of thespatial geometry of the coverage.

In addition, any solution to the problem of represent-ing spatial two-dimensional data in one coverage will haveto fit into a modern storage system that would allow multi-ple viewers and GIS systems to use such a database. It isbecoming clear that this will be an environment that willbe a multi-tiered storage and retrieval geospatial data sys-tem. Most of the proposed systems involve three major

tiers of software. Figure 6 show the OpenGIS approach tothe distribution of GIS data. They are:

1. User software – Viewers, editing, and analysis tools.

2. GIS system software – Arc/Info, AutoDesk Map,Intergraph, SmallWorld

3. Relational database with SQL – Oracle, Xybase,Informix

Translating software may be necessary in betweeneach of the three tiers, using the main software systemsapplication programming interfaces (APIs) shown in figure6, to allow data to move smoothly from one of the tiers toanother. With the advent of such systems and the likeli-hood of GIS data holders moving to this kind of a dataarchive, the spatial data produced by the geologic commu-nity must be in such a form that it will readily be useful insuch an environment. This form includes an understand-ing of the geometric aspects of their products.

CONCLUSIONS

I have digitized and attributed the Cody, WY1:100,000-scale materials to retain the art, esthetics, andgeologic concepts recorded by the map author. I am pro-ducing a map database that when rendered as a map willhave pleasing line work, and yet be useful to geoscienceanalysts. The cleanup and attribution of the map data willtake into account the resolution of the digitizing methodwhile capturing the source data using map accuracy stan-dards, and the concept of map resolution as presented byUCSB as guides. In addition, digital map files for thisproduct will be furnished in a sufficient number of formatsto insure ease of use in a number of common viewers andGIS systems. The attribution of the map and the database


r = 2* grain tolerance

35Scratch Boundary�

(not shown)

w = 4* grain toleranceStart

End

a. "Dot" with a symbol b. "Wire" as a linear object

LeftPolygon

RightPolygon

Figure 5. Point and line objects captured as polygons.Figure 6. Three-tiered data delivery, with their ApplicationsProgramming Interfaces (APIs).

follows the NGMDB data model preliminary standardsversion 4.3 using the “Curly tool” (Raines and Hastings,1999) for the data entry and export into Arc/Info. The pro-posed North American data model standard for geologicmaps is available at: <http://geology.usgs.gov/dm>.

The future of the geologic database may be withArc/Info version 8 on Windows NT systems using objectmodels or with GIS systems like SmallWorld or even bythe use of polygons to represent all geologic informationin current GIS systems. A continuing study of current spa-tial geometry standards by one of the current AASG/USGSworking groups or a new working group is needed to seehow these spatial geometries fit the needs of digital geo-logic maps. Clearly much remains to be done to standard-ize digital geologic map geometry.

REFERENCES CITED

Federal Geographic Data Committee (FGDC), 1998, Contentstandard for digital geospatial metadata (CSDGM):<http://www.fgdc.gov/metadata/contstan.html>.

OpenGIS, 1999,<http://www.opengis.org/techo/presentations/overview/side063.htm>.

OpenGIS, 1998, <http://www.opengis.org/techno/specs.htm>(click on Topic 1).

Pierce, W.G., 1997, Geologic map of the Cody 1-degree by 2-degree quadrangle, northwestern Wyoming: U.S. GeologicalSurvey Miscellaneous Investigations I-2500, scale1:250,000.

Reynolds, Mitchell W., 1971, Geologic map of the Bairoil quad-rangle, Sweetwater and Carbon counties, Wyoming: U.S.Geological Survey GQ-913, scale 1:24,000.

Raines, Gary L., and Hastings, Jordan T., 1999, Curly:Unpublished software tool produced under the auspices ofthe AASG/USGS geologic map data model working group,available at: <http://geology.usgs.gov/dm/tools/Curly>.

US Bureau of the Budget (USBB), 1947, National Map AccuracyStandards:<http://rockyweb.cr.usgs.gov/nmpstds/nmas.html>.

USGS-NMD, 1998, Standards for revised primary series quad-rangle maps (Draft for implementation):<http://rockyweb.cr.usgs.gov/nmpstds/qmapstds.html>.


101

INTRODUCTION

GeoMatter (Geologic Map Attributer) is a prototypedata entry tool that enables the population of databasesstructured in the proposed U.S. national digital geologicmap data model format (Johnson and others, 1998). It wasdeveloped by Geological Survey of Canada to facilitategeologic map data entry as well as explicate the datamodel. Users are expected to have in their possession adigitized version of their geology in ESRI’s shape file for-mat, and an empty (or not) MS-Access database conform-ing to the data model version 4.3, prior to using the soft-ware. GeoMatter provides a graphic environment wherethe geological map and its underlying database coexist on-screen during data entry. Database contents are linked tomap objects and this provides the map not only greaterinformation content but also defines its cartographicappearance. Once linked, the map and database remainsynchronized, such that selecting an item in the databasecauses related map objects to be highlighted, and viceversa. GeoMatter thus both expedites data entry and facil-itates information browsing. It is designed to host multiplemap products within its map database. Its usage shouldresult in digital geologic maps that are complete in termsof cartographic appearance, relatively deep in informationcontent, and compliant with a common data standard. Thiswill aid field mappers in creating digital map products,map compilers in integrating map products, and corporateentities in managing and distributing geologic knowledgein map form.

Software

GeoMatter is a stand-alone product that requires noother software to function. It comes with a standardWindows install and un-install script that manages theaddition and removal of the program. The end-user isresponsible for establishing a connection (via ODBC –Open Database Connectivity) to an appropriate database(e.g. MS-Access) and for making the map layers availableby copying them into an appropriate location. A bare-bones user’s manual is provided to aid its usage. It differsfrom other mapping packages, such as ArcView, in its abil-ity to manage a complex geological database and in coor-dinating maps with it.

GeoMatter follows in the footsteps of other prototypedata entry tools, most notably Curly (Raines and Hastings,1998) and LegendMaker (Sawatzky and Raines, 1998). Itdiffers from these approaches in three main ways: firstly,in the coexistence and tight integration of the map anddatabase; secondly, in its user interface which insulates theuser from the underlying database structure; and thirdly, inits inability to adapt to major design changes in the datamodel without additional programming.

Hardware

GeoMatter operates on standard Windows 95 or NTcomputers. In its current state, disk related operations areparticularly slow, and require at least a powerful PentiumII microprocessor and large quantities of memory (e.g. at

Geomatter: A Map-Oriented Software Tool for AttributingGeologic Map Information According to the Proposed

U.S. National Digital Geologic Map Data Model

By Boyan Brodaric1, Eric Boisvert2, and Kathleen Lauziere2

1Geological Survey of Canada615 Booth St.

Ottawa, ON K1A 0E9Telephone: (613) 992-3562


2Geological Survey of Canada2535 Laurier BoulevardSte. Foy, PQ G1V 4C7

least 64Mb) for adequate performance. These and othershortcomings will be remedied in a forthcoming revision.

Distribution

GeoMatter is currently at a prototype stage due totechnical development issues and, in part, due to theevolving nature of the data model. Its main functions areoperational but with much embedded quirkiness. It hastherefore not yet been applied within a serious map compi-lation project, and this will likely remain to be the caseuntil the currently planned revisions are completed.Experimentation with the product is currently limited to part-ners within the data model effort who contact one of theauthors for ‘as is’ access to the software and manual.

Future Directions

GeoMatter is currently being revised and completedfor production usage. Many current deficiencies are beingcorrected and some necessary functionality added, such asfull compliance with v4.3 of the proposed U.S. nationalgeologic map data model. It is anticipated that this workwill be complete in fall/winter 1999.

The discussion below touches on some historicalaspects of GIS software development that impacted thecreation of GeoMatter, followed by a description of theuser interface and its interaction with the proposed U.S.national digital geologic map data model.

DATA MODELS AND THE NEED FORAPPLICATION SOFTWARE

The establishment of a standard data model for digitalgeologic map information is a critical component of thenational map database initiatives in the US (Soller andBerg, 1997), Canada (Broome and others, 1998), and else-where (Bain and Giles, 1997; Ryburn and O’Donnell,1998). Designers of these data models are faced with adaunting task in that the complexity of geological informa-tion requires equally complex database architectures: it isinconceivable to imagine that the computer representationof a complex earth system would be significantly simplerwithout great loss of context, content and, ultimately, utili-ty. Mature data model efforts in the geosciences (POSC,1998; PPDM, 1998) reflect this reality, being composed ofhundreds of interrelated parts, such as tables in relationaldatabases or objects in object-oriented systems. Oftenthese parts do not reflect actual geologic elements but areartifacts of the computing system being used. It is there-fore understandable why end-users feel lost in the maze ofone data model or another and are reluctant or unable topopulate these complex structures that seem foreign andremoved from their view of the world. In database tech-nology this conflict between how reality is represented by

the computer versus the user’s perspective has been longaddressed and solved: application software presents thedatabase to the user in distinct views (Date, 1990) thatcoincide with the user’s conceptions, and the underlyingdata structures remaining hidden. Intermediary software isthus expected to mediate between user and database. Toaid this process the database community has evolvedCASE (Computer Assisted Software Engineering) toolsthat greatly ease the construction of application softwarefrom initial database design. Indeed, these are so maturethat the database and application software design processesseem as one task within such environments, as the toolsgenerate both a database structure and accompanyingapplication software.

In GIS however, we are only now beginning to see theemergence of sophisticated tools for application softwaredevelopment. Partial responsibility for this lies in GIS’relative immaturity; however, it is also probably due to thefact that GIS explicitly deals with the added complexity ofspatial information which possesses no unifying theory(Edwards, 1996; Frank and Kuhn, 1995), leading to dis-parate, often conflicting, implementations (Gahegan,1996). Although several ad hoc data exchange standardssuch as DXF (AutoDesk, 1999) and government led effortssuch as SDTS (USGS, 1994) and SAIF (Canadian GeneralStandards Board, 1995) have come into existence, they arenot theories of spatial computation in the same sense thatthe relational data model defines relational data manage-ment or that object-orientation defines object computing.Emerging spatial standards (Open GIS, 1998; ISO/TC 211,1998) improve this situation to a degree, but they are alsoimmature and, arguably, incomplete, and have not yetengendered viable software application environments. Onthe other hand, the explosion in the usage of object-orient-ed programming, and its subsequent recent adaptation bythe GIS community, has caused individual vendors toforge onwards and create spatial tool kits that are powerfulalbeit diverse. The challenge in developing effective spa-tial database software now rests in sifting these variousvendor offerings for a suitable match to the problem athand.

DESIGN OF GEOMATTER

The proposed U.S. national digital geologic map datamodel is at once both a conceptual and logical representa-tion of a geologic map. It is conceptual in that the funda-mental components of a geologic map (as understood bythe designers) have been abstracted for computer represen-tation independent of any database implementation. Thisis expressed primarily in the descriptive text of Johnsonand others (1998). The remaining text and diagrams of thereport are oriented to a relational database description ofthese concepts and in this way constitute a logical, tech-nology-specific, expression of them. When the logical


design is applied within a specific relational database sys-tem, such as Oracle or Access, the resultant configurationrepresents a physical model.

For the software designer these distinctions can bevery useful, and can in fact become the basis for thedesign of application software. The conceptual elementsthat comprise a map (e.g. legend, text, map features, etc.),being user-oriented, provides a good basis for a user-inter-face, whereas the logical and physical aspects (i.e. the rela-tional database design) provide a foundation for the under-lying data repository. For instance, if software designerschoose to adopt SQL (Standard Query Language) as thelanguage for manipulating relational databases in their pro-gramming, then they have in effect enabled their softwareto operate with any database that is SQL compliant. Thisillustrates the logical-physical division, as the logical SQLpart is syntactic and unchanged across hardware environ-ments, whereas the databases that respond to SQL mustdiffer internally to cope with different physical hardwareplatforms—thankfully, in a way that is largely transparentto the user.

From this we can envision a software tool that pre-sents a geological user interface (conceptually), communi-cates to its data store via SQL (logically), and permits con-nection to a variety of relational data repositories (physi-cally). This is essentially the design of GeoMatter. Inexecuting this design, the Delphi object-oriented program-ming environment was used to enable rapid developmentof the user interface and to facilitate its connection to theunderlying relational databases via SQL and ODBC;ESRI’s MapObjects component (ActiveX) was used formap display, symbolization and various spatial operations.

USER INTERFACE

Following the data model design, the user interface isorganized around six main concepts: sources, legend, rockunit, structural unit, occurrence and map (spatial objects).GeoMatter permits relevant information to be added tothese components and enables linkages between them tobe established. It is designed to accomplish this in a waythat is familiar to the geologist by mimicking the tradition-al map construction process. The map is drawn on one ormore layers that host geological boundaries and associatedgeologic features. These are given meaning by relatinggeological information to them through a legend that alsodetermines their cartographic appearance. Hence the maptakes shape by defining a map Legend which containssymbolized Rock Units, Structural Units or Occurrences,which in turn are then connected to a map description inthe Sources and to specific map features on the map face.

These activities take place in two main windows(Figure 1): an information window on the left, where thetext information is managed, and a map window on theright where the map is displayed. A geologist can there-

fore maintain the map display and simultaneously browse,add and edit information about specific map features.Moreover, focus is maintained throughout, such that as aspecific item is chosen from the map or from the database,all related map and database items are selected and high-lighted, thus enabling the geologist to quickly view interre-lationships among them. For example, if an item in thelegend is selected by the user then its related map objectsare highlighted on the map, and its associated map defini-tion, rock units, structural units or occurrences are selectedand displayed within the individual information compo-nents.

Sources

The Source component (formally known as Metadatain v4.2—Johnson and others, 1997) is a repository for doc-ument descriptions that are cited in other parts of the data-base. These could be maps and reports or any other docu-ment that needs to be referenced within the data model.Sources can be browsed, added and deleted here.Selecting a source that has a legend associated to it willcause the map to be symbolized according to the attachedlegend. This capability has interesting ramifications as itallows derivative map products to be created and stored, asseveral legends, and thus many map sources could be asso-ciated with any one set of geographic layers. For instance,the spatial objects in the map window could be representedas a bedrock geology map, a tectonic unit map, a dominantlithology map, etc., each having a unique legend in theLegend section and unique reference entry in the Sourcesection. Viewing the objects in the map window accordingto any one theme simply requires the appropriate referenceto be selected. Since the various sections are linked,browsing the data contents thereafter will result in thehighlighting of information specific to the selected theme(e.g. tectonic units or lithologies instead of bedrock units).The ability to display different map layers in the map win-dow, permits users to manage multiple map products fromone database. Note that map layers are only displayableand not editable – GeoMatter is not a digitizing tool anddoes not currently allow map features to be modified,though this might change in the next version.

Legend

In many ways the Legend component is the heart ofGeoMatter. It effectively acts as a visualization device,coordinating specific data content, symbolization, and spa-tial objects to form a ‘map’. Thus, although in appearanceit resembles a typical geologic legend, it in fact best illu-minates a nontraditional implication of the data model:that a map is simply a specific view of a geologic data-base, defined by the collection of specific symbolic, the-matic and spatial parts. It is thus possible to rearrangethese components and arrive at a different map from the

GEOMATTER: A MAP-ORIENTED SOFTWARE TOOL FOR ATTRIBUTING GEOLOGIC MAP INFORMATION 103

same pool of data. This amounts to classifying a set ofspatial features according to different Legends. As aLegend can only be related to a single map Source,GeoMatter requires a new map Source to be definedbefore another Legend can be ascribed to the same mapfeatures.

A Legend is displayed as a list of individual items(Figure 2), each containing a symbolization and thematicdescription. The Legend provides efficiencies in editingand browsing map data. Individual map objects can beselected from the map and assigned to a legend item. Inbrowse mode, selecting a map object will cause its legenditem to be highlighted, or vice versa, the selection of a leg-end item will cause all associated map objects to be high-lighted, making very clear the distribution of that item onthe map.

Rock Units and Structural Units

The geological data model differentiates between cate-gories of geologic information (COA — Compound ObjectArchive) and specific instances of geologic occurrences(SOA — Singular Object Archive). For example, a specif-ic structural measurement occurrence has a unique identity


Figure 1. The GeoMatter interface — an extensible database window for geologic information is situat-ed on the left, and a window for displaying the map layers is found on the right. The database window isdisplaying the Source component, which lists the available maps and permits their selection and modifi-cation. Once a map is selected, the map layers acquire cartographic and geologic characteristics.

Figure 2. Part of a Legend. Each legend itemcontains a symbol, map label, text description,and a link to a geologic category. Objects on themap can be assigned to a legend item, and willthereby inherit the cartographic and descriptivecharacteristics represented by the item.

and defining characteristics (e.g. position, strike, dip), butbelongs to a general category (e.g. foliation); likewise‘fault’ is a generic category whereas the ‘Logan fault’ is aspecific instance of the general category. This is veryanalogous to the geological mapping process where thegeologist abstracts a generalized category, call it ‘mapunit’, from a set of descriptions located at specific occur-rences.

The Rock Unit and Structural Unit components repre-sent the list of available categories within the system.Rock Units typically characterize polygons and StructuralUnits linear features, but this is user-driven as GeoMatterwill allow any geometric shape to be attached to a rockunit, structural unit or occurrence. The category lists maybe arranged hierarchically (Figure 3a) to follow geologicalrelationships: e.g., a formation contains members and maybe contained within a group. Mechanisms are provided to

modify the hierarchy (Figure 3b) and to alter the level ofdetail displayed. Individual categories may be furtherdescribed according to their rank (for rock units),geochronologic age, stratigraphic age, rock compositionand their relationship to other units (e.g. overlying, con-temporaneous, etc.). Specific hierarchic lists are incorpo-rated at the appropriate places during data entry, to controlthe content of certain database items, such as rock type ter-minology. As with the Legend component, selecting a unitwill highlight its occurrences on the map, and vice versa,selecting a map object will highlight its associated catego-ry in the units’ list.

Occurrences

Fossils, structural measurements, mineral deposits,and individually named plutons are examples of occur-rences that can be described by GeoMatter. In addition,specific map units (e.g., polygons) can be assigned propor-tional compositions of lithologies or other map units.Again, this follows the mapping process where a generalcategory is a best fit description of the occurrences leadingto its inception; unless these occurrences represent a typelocality for the category they will likely differ somewhatfrom it. For example, a map unit may contain mostlyshale, then silt and sandstone, but a specific occurrence(e.g., a polygon) may contain only a subset of these invarying proportions.

GEOMATTER: A MAP-ORIENTED SOFTWARE TOOL FOR ATTRIBUTING GEOLOGIC MAP INFORMATION

Figure 3a. The rock unit tab contains a hierarchy ofrock unit categories and their lithologic composition andage descriptions.

Figure 3b. Modifying the rock unit hierarchy. Rock unitcategories can be added, deleted or moved.

105


CONCLUSIONS

GeoMatter could be used by any geologist interestedin boosting the information content of their existing digitalgeologic map. It is designed to enable the assignation ofcartographic and geologic attributes to digitized points,lines or polygons, and to browse resulting map databasecontents. This should enhance the information content ofdigital geologic maps and thus aid the usefulness of indi-vidual map products and promote the construction ofbroader map databases whose contents are interchange-able. It should particularly benefit map compilers inter-ested in facilitating the integration of diverse map productsby standardizing on one database structure. The lattermight include not only individuals but also organizationsdeveloping corporate map databases for internal usage orexternal distribution.

Underlying all such activity must be the belief thatscientific understanding will be advanced through the syn-thesis of information that was previously difficult to inte-grate. Developing effective tools that will aid the popula-tion of interchangeable databases becomes imperative tothis effort, and it is in this regard that GeoMatter hopes tomake an impact. At the moment it is a prototype effortthat will soon be upgraded to a system capable of beingutilized in production mode.

REFERENCESAutodesk Inc., 1999, DXF Reference,

<http://www.autodesk.com/techpubs/autocad/dxf>.

Bain, K.A., and Giles, J.R.A., 1997, A standard model for storageof geological map data: Computers and Geosciences, vol.23, no. 6, p. 613-620.

Broome, J., Kilby, W., and Denoyers, D., 1998, Developing theCanadian Geoscience Knowledge Network: A DistributedApproach, in Buccianti, A., Nardi., G., and Potenza, R(eds.), Proceedings of the Fourth Annual Conference of theInternational Association for Mathematical Geology, Part 2:De Frede, Naples, p. 663.

Canadian General Standards Board, 1995, Canadian GeomaticsInterchange Standard (CAN/CGSB-171.1-95) or SpatialArchive and Interchange Format (SAIF): Canadian GeneralStandards Board, Formal Definition, Release 3.2, January1995, 258 p.

Date, C.J., 1990, An Introduction to Database Systems, Volume1, Fifth Edition: Addison-Wesley, New York, NY, 854 p.

Edwards, G., 1996, Geocognostics - A New Paradigm for SpatialInformation: Proceedings AAAI Spring Symposium, March25-27, 1996, Stanford University.

Frank, A.U. and Kuhn, W., 1995, Specifying Open GIS withFunctional Languages, in Max J. Egenhofer, and John R.Herring (eds.), Advances in Spatial Databases (SSD’95):Springer-Verlag, Lecture Notes in Computer Science, Vol.951, pp. 184-195.

Gahegan, M. N. 1996, Specifying the transformations within andbetween geographic data models: Transactions in GIS, Vol.1, No. 2, pp.137-152.

Johnson, B.R., Brodaric, B., Raines, G.L., Hastings, J., and Wahl,R., 1998, Digital Geological Map Data Model v4.3.<http://ncgmp.usgs.gov/ngmdbproject>.

Johnson, B.R., Brodaric, B., and Raines, G.L., 1997, DigitalGeological Map Data Model v4.2, 83 p.<http://ncgmp.usgs.gov/ngmdbproject>.

ISO/TC 211/WG2, Geographic Information: Geospatial DataModels and Operators,<http://www.statkart.no/isotc211/wg2/wg2welc.htm>.

Open GIS Consortium Technical Committee, 1998, The OpenGIS Guide, Third Edition, Buehler, K., and McKee, L.(Eds.): Open GIS Consortium, Wayland, MA,<http://www.opengis.org/techo/guide.htm>.

POSC, 1998, Petrotechnical Open Software Corporation,<http://www.posc.org>.

PPDM, 1998, Public Petroleum Data Model Association,<http://www.ppdm.org>.

Soller, D., and Berg, T.M., 1997, Progress Toward Developmentof the National Geologic Map Database, in Soller, D.R.(ed.) Proceedings of a workshop on digital mapping tech-niques: methods for geologic map data capture, manage-ment and publication: U.S. Geological Survey Open FileReport 97-269, pp. 37-40.<http://ncgmp.usgs.gov/pubs/of97-269/>.

Raines, G. L., and Hastings, J.T., 1998, Curly and Curly SurvivalGuide. <http://geology.usgs.gov/dm/tools/Curly/>.

Ryburn, R., and O’Donnell, I., 1998, Towards national geo-science data standards: Australian Geoscience ResearchNewletter 29, November 1998,<http://www.agso.gov.au/information/publications/resnews/1998>.

Sawatzky, D.L. and Raines, G.L., LegendMaker andLegendMaker Guide,<http://geology.usgs.gov/dm/tools/LegendMaker/>.

USGS, 1994, Spatial Transfer Data Standard,<http://rmmcweb.cr.usgs.gov/public/nmpstds/sdts.html>.

INTRODUCTION

The Kansas Geologic Names Database (NDB) is arelational database of stratigraphic nomenclature. It hasevolved, with extensive modification, from the text files ofthe recently published Lexicon of Geologic Names ofKansas (through 1995), edited by Baars and Maples(1998). The NDB development process has proved to be avaluable and practical aid toward implementing the digitalgeologic map data model (MDM) proposed by Johnson,Brodaric, and Raines (1998) through the NationalGeologic Map Database (NGMDB) Project. In particular,the nomenclature database facilitates development of thetables for Kansas related to rock unit definitions within theMDM. It is applicable to stratigraphic nomenclatureappropriate to geologic maps and other reports on thegeology of Kansas published currently or at any time inthe past. When dealing with previously published maps,the nomenclature database links rock unit names in use atthe time of publication with both prior and subsequentvariations in accepted nomenclature, and with the sourcesof those changes. This paper reviews development of thenomenclature database and its associated data model, theinfluence of this project on development of a geologic mapdata model for Kansas, and variations from the data modelof the NGMDB Project.

OBJECTIVES AND MOTIVATION

The primary goal of the nomenclature database projectis the organization and enhancement of information fromthe Kansas Lexicon into a form that permits easy access to

descriptive information on rock units based on a wide andflexible range of user needs and selection criteria. Onegeologist may be interested in the current nomenclature ofthe formations within the Admire Group (UpperPennsylvanian, Virgilian Series), along with the locations,descriptions, and available images of their type sections.Another geologist’s interest may focus on the acceptednomenclature for the Admire Group prior to 1938 (thenconsidered Permian) and its relationship to the previouslydefined “Admire shales.” Someone studying the history ofgeologic research in Kansas might want a list of all rockunits first described in publications having R. C. Moore (aformer state geologist) as principal author.

Numerous other goals are tied to flexible access torock unit information. It is clearly desirable to linkdescriptive text files for a specific rock unit to correspond-ing digital map objects, photographs, or document images.On-line edit capabilities permit easy correction of errorsfound within the nomenclature database, with immediatedisplay of corrections to users. A capability for publica-tion on-demand directly from the database permits prompt,cost-effective publication of enhanced or specialized lexi-cons as information within the database is improved.

Geologic mapping, broadly defined, is the fundamen-tal data collection and information management activity ofthe geologist. With the rapid pace of digital geologic mapdata development in Kansas, the nomenclature database isneeded for direct support of mapping and related publica-tion activities. The importance of the nomenclature data-base for implementation of the NGMDB Project’s pro-posed geologic map data model standards became apparentin the early stages of the project, with recognition that theproposed standards had much to offer toward design of the

107

Development of the Kansas Geologic Names Database: A Linkto Implementing the Geologic Map Data Model Standards

By David R. Collins and Kurt Look

Kansas Geological SurveyCampus West, University of Kansas

1930 Constant AvenueLawrence, KS 66047

Telephone: (785) 864-3965Fax: (785) 864-5317

e-mail: [email protected]@kgs.ukans.edu

nomenclature database. Both efforts are viewed as a steptoward development of a general model for all geologicdata, as suggested by Richard (1998).

Many of these objectives arose in direct response toshortcomings of the Lexicon, which was developed in aword processing environment. The Lexicon’s digital textfiles lack organized database structure and are not publiclyaccessible. By its nature, a lexicon of stratigraphicnomenclature is characterized by repetitive use of a limitedset of information types. Contributions to the Lexiconfrom numerous stratigraphers, combined with the largenumber of included names, contributed to inconsistentstyle, format, and information content for named units.Use of abstracts from earlier lexicons perpetuated previ-ously published errors. This practice also resulted in fre-quent occurrences of citations with imbedded references tosources described by author, date, and page with no furtheridentifying information to be found anywhere in theLexicon. In its printed form, the Lexicon provides no visu-alization of type sections or the geographic extent ofnamed units. Direct support to digital mapping activitiesis not practical with the structure of the Lexicon’s textfiles. In an effort to conserve space and limit the publica-tion to a manageable size, a considerable amount of usefulinformation was excluded from the Lexicon.

THE NDB DEVELOPMENT PROCESS

Given the objectives of the geologic names databaseproject, relational database capabilities clearly offered apractical means of addressing the task at hand. In a criti-cal first step, the original Lexicon text files were reformat-ted to facilitate parsing into separate fields. Significantrevisions and enhancements occurred as information con-tained in the original files was checked for errors andomissions.

As development proceeded, it became apparent thatsimilar data structures were appropriate for management ofdigital geologic map data and for management of historicalinformation about the names of geologic rock units.Information on digital geologic map data models availablethrough the NGMDB Project web-site<http://ncgmp.usgs.gov/ngmdbproject/> simplified thedevelopment process. It provided a clear starting point,identifying critical tables, data fields, and relations.

An iterative process was implemented to achieve abalance between the additional effort of working aroundmore problems in the full text files and the additionalgains from further identification of useful text segmentsprior to parsing and loading into the relational database.Once that balance was reached and parsing routines werethoroughly tested, the information from the enhanced andreformatted Lexicon files was parsed and loaded into therelational database management system. A large portion ofthe information went into three tables of the new NDB; theSOURCES, CITATIONS, and ROCK_UNITS tables(Figure 1). There are one-to-many links from SOURCESto CITATIONS (each source may have citations relating tomany different rock units) and many-to-one links fromCITATIONS to ROCK_UNITS (where citations frommany sources may define a particular rock unit). Furtherparsing, as needed, will be done within the relational data-base management system.

The SOURCES table contains a separate record foreach unique information source. There are 914 sourcescurrently identified in the NDB. Records contain basicbibliographic information, source format (book, journal,note, map, etc.), and (where appropriate for specific geo-logic reports) information on the geographic extent of thestudy. A recursive source relationship (“contained in”) isbuilt into the SOURCES table. One source may containmany other sources. For example, an issue of a journalmay contain many articles. Currently, 186 sources areidentified as “contained in” 97 of the other sources.Records within the ROCK_UNITS table provide the basicidentifying information for each recognized geologic rockunit name. This includes name, name origin, lithostrati-graphic or chronostratigraphic rank, the names of each unitof higher rank containing the original unit, text statementsof geographic extent and pointers to map objects for visu-alization of geographic extent. There are 1820 unit namesin the database, including about 250 chronostratigraphicnames and 1570 lithostratigraphic names. Fewer than 500of the lithostratigraphic names are currently accepted asformal names in Kansas. A recursive unit relationship(“current_usage”) is built into the ROCK_UNITS table tolink abandoned unit names to the currently acceptednomenclature for the corresponding unit. Each record ofthe CITATIONS table, linking SOURCES andROCK_UNITS, contains descriptions or comments regard-ing a specific rock unit, obtained from a specific source,


Figure 1. Primary tables in the NDB, reflecting the interface of information sources and definedrock units with specific citations.

with reference to the specific location of the informationwithin that source. There are 5179 separate citations; anaverage of 2.8 citations per named rock unit.

The overall structure of the digital geologic map datamodel (MDM), as presented in Figure 2-6 of Version 4.3(Johnson, Brodaric, Raines, Hastings, and Wahl, 1998, p.7), is generalized here in Figure 2. The model has fourmajor components. The METADATA section providesdetailed information about the information sources (i.e.,data about data). The SOURCE table is the primary tablewithin the metadata section of the MDM. In the MDM thesources are typically either published maps, sources con-taining the published maps, or the documents and databas-es from which the published maps were derived. TheCOMPOUND OBJECT ARCHIVE of the MDM providesdata structures for information related to all complex geo-logic features found in the real world, including a rock unittable and related descriptive tables. The SPATIALOBJECT ARCHIVE section maintains data on map

objects used in visualizations of particular rock units. TheLEGEND, with associated classification schemes, providesthe functional details for specific map visualizationsachieved through the combination of spatial object repre-sentations of rock units.

A more detailed view of the metadata portion of theNDB is provided in Figure 3. These tables provide infor-mation collectively describing the origins and nature ofavailable geologic information. They correspond to themetadata tables of Version 4.3 of the MDM.

A separate AUTHORS table has been added to facili-tate access to work by particular authors in the NDB. Anintersection table (X_AUTHORS_SOURCES) linksauthors to each of their publications, identifies theirsequence in a list of contributing authors, and links theauthor to their employing organization for that publication.Sources are linked to publishing and funding organizationsthrough a separate intersection table (X_SOURCE_SUP-PORT). This permits many-to-many relationships betweenfunding agencies and information sources, and betweenpublishers and information sources, to be handled as com-pound one-to-many relationships. The SOURCES_RELA-TIONSHIPS table links sources within the SOURCEStable through relationships such as “complies with [thespecified standard]” or “digitized from [the specifiedsource]” as defined in a data dictionary. The PROJEC-TIONS table provides the additional information unique toinformation sources with map formats.

Details of the portion of the NDB corresponding to acompound objects archive are seen in Figure 4. The“Formal Unit” and “Rock Unit” tables of Version 4.3 ofthe MDM are merged into a single ROCK_UNITS table inthe NDB. Sequences of units appropriate to a particularmap, or published in a specific source as “formal” units at

DEVELOPMENT OF THE KANSAS GEOLOGIC NAMES DATABASE 109

Figure 2. Generalized structure of Version 4.3 of the dig-ital geologic map data model (Johnson, Brodaric, Raines,Hastings, and Wahl, 1998).

Figure 3. Metadata tables in the NDB, describing the origins and nature of available information.

a particular time, are listed in the SEQUENCE table.Records in the OCCURRENCES table give specific loca-tions where a rock unit has been described, ranging fromthe defining holostratotype (original type section) of a rockunit to a local measured section that includes all or part ofthe unit. The OCCURRENCES_IMAGES table providespointers to map objects, digital photographs, or scannedrecords (such as the published type section or a measuredsection) related to an occurrence of a rock unit. TheOCCURRENCES_SECTIONS table is presented here inplace of the full range of descriptive tables found inVersion 4.3 the MDM. Searchable data related to litholo-gy, composition, fossil assemblages, thickness and otherclassifying characteristics of a rock unit would be foundhere, separate from general descriptive statements found inthe CITATIONS table. The characteristics found under theOCCURRENCES_SECTIONS table and the relationshipsfound in the UNITS_RELATIONSHIPS table are definedin data dictionaries covering the full range of relationships(hierarchy, classification, correspondence, proportion, anddisposition) described by Richard (1998).

STATUSThe design of the Kansas Geologic Names Database is

consistent with the corresponding elements of the pro-posed geologic map data model standards, and represents amajor step toward full implementation of those standards.Functions defined in tables in the LEGEND portion ofVersion 4.3 of the MDM (see Figure 2), control productionof visualizations of geologic map data. Similar functionsare defined for the NDB using commercial report writer

software (available either as components of the relationaldatabase management system, or as separate systems) tocontrol report generation for a complete and current lexi-con of geologic names in Kansas or selected subsets. Forexample, a lexicon could be extracted from the databasefor geologic names used by Moore, Jewett, and O’Connor(1951) in their geologic map of Chase County, Kansas.

Universal web access is now under development. Arevised lexicon of geologic names in Kansas will be pub-lished on-line at the Kansas Geological Survey’s web site<www.kgs.ukans.edu>. The on-line publication willaccompany a web conference site of the KansasNomenclature Committee for discussion of nomenclatureissues, contributions of new information, and reporting oferrors within the database. This will be similar to the webconference site used for discussion of the geologic mapdata model standards by the AASG/USGS Geologic MapData Model Working Group at <http://geology.usgs.gov/dm/>.

Merging the NDB with the MDM will result in a geo-logic data model with sources (for particular nomenclaturecitations) as attributes of spatial objects used to representspecific rock units within a particular visualization ofregional geology. The concept of a geologic names data-base can be broadened to include the historical develop-ment of accepted names for specific occurrences of struc-tures or other geologic features in addition to rock units.

CONCLUSIONS

(1) The high degree of effort required to publish completelexicons of geologic names by the traditional printing


Figure 4. Rock unit tables in the NDB, providing descriptions and relationships ofrock units.

process, accommodating a relatively small proportionof new or revised names, has made such tasks a lowpriority for most geologists.

(2) Consistent formats throughout large printed volumes(almost impossible to achieve as a manual process andstill not easily obtained using word processing soft-ware) become feasible in a relational database envi-ronment.

(3) Books, including lexicons, are just like published geo-logic maps — you always discover important omis-sions and uncorrected errors after they go to press.The larger the press run, it seems, the more numerousand significant the errors.

(4) On-demand publication and distribution from relation-al databases provides a more efficient and cost-effec-tive method for geological surveys to maintain formallexicons and provide access to information on geolog-ic nomenclature.

(5) Universal, on-line, access provides strong incentivesfor geologists to participate in the contribution of newinformation or identification of errors within the data-base by limiting their involvement to productive activ-

ities and providing rapid incorporation of contribu-tions into the public domain.

REFERENCES

Baars, D. L. and C. G. Maples, eds., 1998, Lexicon of GeologicNames of Kansas (through 1995): Kansas GeologicalSurvey, Bulletin 231, 271 p.

Johnson, B. R., B. Brodaric, and G. L. Raines, 1998, Digital geo-logic map data model; version 4.2: AASG/USGS GeologicMap Data Model Working Group,<http://ncgmp.usgs.gov/ngmdbproject/standards/datamodel/model42.pdf>.

Johnson, B. R., B. Brodaric, G. L. Raines, J. T. Hastings, and R.Wahl, 1998, Digital geologic map data model, Addendum toChapter 2; version 4.3: AASG/USGS Geologic Map DataModel Working Group,<http://geology.usgs.gov/dm/model/Chapter2add.pdf>.

Moore, R. C., J. M. Jewett, and H. G. O’Connor, 1951, ArealGeology of Chase County, Kansas; in, Geology, MineralResources, and Groundwater Resources of Chase County,Kansas: Kansas Geological Survey, Volume 11.

Richard, S. M., 1998, Digital Geologic Database Model: ArizonaGeological Survey, <http://www.azgs.state.az.us/GeoData_model.pdf>.

111

INTRODUCTION

In 1995, the use of computer technology for makingmaps was increasing in the Mines and Minerals Divisionof the Ontario Ministry of Northern Development andMines. In addition, the Ministry had, by this time,embarked on a massive data conversion project known asthe Earth Resources and Land Information System(ERLIS) to make information available on a central sys-tem. Different GIS solutions were adopted and imple-mented throughout the Ministry resulting in a plethora ofapplications, each with a set of characteristics that met theneeds of each section. The cartographic unit of thePublication Services Section (PSS) had migrated from thepaper world to using Intergraph as its mainstay in mapproduction. The Data Services Section (DSS) built ERLISaround Genasys Genamap GIS and Oracle DBMS. TheOntario Geological Survey (OGS), a branch of theMinistry of the Northern Development and Mines, hadalso implemented various software tools. ThePrecambrian Geoscience Section (PGS) used AutoCADand Fieldlog as their data entry and map creation software.The Sedimentary Geoscience Section (SGS) usedMicrostation as their key, drafting tool. Together the PGSand SGS used IDRISI and MapInfo for GIS related work.Despite the variety of applications, the Ministry created anefficient cartographic process that generated quality, hardcopy color maps in an on-demand format.

With the initiation of the Abitibi Compilation Projectthe OGS decided to assess the usefulness of a full-blownGIS package capable of integration, compilation, analysesand exchange. To assess the degree to which this func-tionality was required, the OGS embarked on a“Functional Analysis” under the guidance of the DataServices Section. Following the Functional Analysis, aformal evaluation of several GIS platforms was completed

and a GIS system that best satisfied the OGS InformationTechnology requirements was purchased.

Using the procedures developed in the FunctionalAnalysis, compilation of the Timmins sheet took place andboth a paper colored map and digital product were pub-lished in March of 1998. This product represents sheetone of a four-sheet product with the remainder slated forrelease by 2000.

GEOLOGICAL OVERVIEW

The Abitibi Subprovince is an 800 by 300 km Archean“granite-greenstone” domain. The value of mineral pro-duction from this area is the highest in the province,approaching 1 billion dollars (Cdn.) (Thurston 1996). It isdominated by supracrustal and granitoid rocks that rangefrom 2.67 to 2.75 Ga (Jackson and Fyon 1991). TheTimmins map sheet covers an area of approximately 9000sq. km., centered on the Timmins mining camp. Rocksare classified on the basis of their dominant lithologyusing textures, structures and composition. Preliminarygeological information was compiled from previous map-ping. New interpretations of the extent of lithologicalunits, specifically in the areas lacking outcrop, greatly ben-efitted from the use of the reprocessed geophysical data(Gupta 1995, 1996). As well, geochemical data allowedfor further subdivision of the geological stratigraphy.

Within the confines of the Timmins map sheet lies thePorcupine mining camp, one of the preeminent lode goldmining districts in the world. Significant base metal pro-duction has also come from this area, mainly from theKidd Creek deposit. Komatiite-associated nickel depositshave been mined intermittently. Non-metallic mineralssuch as scheelite, asbestos and talc have also been extracted.

113

From Functional Analysis to CD — Digital Compilation of theTimmins Map Sheet, Abitibi Greenstone Belt, Ontario

By Brian Berdusco, N.F. Trowell, J.A. Ayer, Z. Madon, S. van Haaften, and A. Yeung

Ontario Geological SurveyB7-933 Ramsey Lake RoadSudbury, Ontario P3P 1X7Telephone: (705) 670-5725


STANDARDS

This digital cartographic process required the develop-ment of symbol libraries. With these symbol libraries, theOGS has been able to provide clients with a common lookand feel to published maps. To date, the OGS has devel-oped a library of over 1600 bedrock mapping symbols thatrepresent features such as sedimentary and volcanic bed-ding, layering, unconformities, structural features, etc.(Jackson et al, 1995; Muir 1995).

Concurrent with the Abitibi Compilation Project, theOGS continued development of digital line standards, min-eral deposit symbol standards and symbol standardsreflecting zones of alteration and deformation (stipple pat-terns). Other libraries under consideration include symbol-ogy for metallogenic classification, Quaternary geologyand mineral commodity classification. The developmentprocess for each symbol library requires considerableeffort. In an effort to decrease development time, the OGShas requested symbol libraries from other provincial, stateand federal surveys.

To make effective use of symbol libraries in our GIScompilation, it is necessary to port existing libraries overto the GIS environment. Symbol libraries up to this pointhad only been developed for the CAD and cartographicenvironment. Off the shelf font creation software wasused to create comparable symbols for the GIS environ-ment. Symbols were converted on an “as required” basis.Eventually, symbol libraries will exist in both CAD andGIS formats. Pending the publication of standard OGSsymbology for mineral deposits, the Abitibi CompilationGroup adopted and modified the GSC mineral depositsymbology (Eckstrand et al, 1995). This symbology,though acceptable, does not classify based on metallogenicprocesses, a desired requirement of the OGS.

Digital line standards have been developed to reflectfolds, faults and contacts. These standards will be incor-porated into further releases and the final release of GISproducts by the Abitibi Compilation Project.

BUSINESS FUNCTIONAL ANALYSIS

The use of digital technology and geographic informa-tion systems (GIS) for geological compilation requiresmore than the purchase and installation of computer sys-tems. In order to be successful, the adoption of new tech-nology must be carried out with a new perspective of, andmore importantly, a new approach to doing business. Inthe case of the Abitibi compilation, this implies thatinstead of treating the compilation as a conventional map-ping exercise, we have to look at it from the perspective ofspatial information management. In other words, the taskmust be completed using the concepts and techniques fordatabase and information system development, rather than

as a sequence of loosely connected assignments in a tradi-tional geological compilation project.

Database development usually starts with the processof a business functional analysis. The objective of theanalysis is to develop a business model of the organization(e.g. the OGS) or the specific function at hand (e.g. theAbitibi compilation). This business model depicts all theprocesses involved in completing the business functionand is used as the basis for the following objectives:

(1) to identify the information products (maps, graphs,charts and written reports) produced in the businessfunction;

(2) to identify the data required to produce the informationproducts in (1);

(3) to find the sources of data identified in (2); and

(4) to develop a strategy and procedures to evaluate thesuitability of the data in (3) if they are available; or

(5) to develop a strategy and procedures to collect newdata to meet the requirements of the data in (3).

There are various ways by which the businessprocesses can be identified. The most commonly usedmethod is to adopt a top-down approach in which businessprocesses are identified and continuously decomposeduntil individual processes can be completed in one singleactivity or step. The result of this phase of the analysis isdocumented in the form of a Business FunctionalHierarchy. An accompanying document, called FunctionDefinitions, is then developed. The purpose of theFunction Definitions is to explain in detail each of theprocesses (Figure 1). These two documents are used in thenext phase of analysis to identify the information productsgenerated by the business processes, which are document-ed in a Function-Information Product Table (Figure 1,Table 1). In the next phase of the analysis, the sources ofthe data are identified and documented in an InformationProduct-Data Table (Table 2).

In the field of database management, there are differ-ent ways by which business functional analysis can be car-ried out. A popular approach appears to be the JointApplication Design (JAD) methodology (Figure 2), origi-nally developed by IBM as the corporate standard for sys-tems development. This method is based on the core team(also known as focus group) approach. A core team ismade up of a facilitator, usually a systems analyst, and agroup of five to seven members who are very familiar withthe operation of the business. Other members can becoopted from time to time. These include: (1) professionaland technical staff who are familiar with the operation ofthe business but do not have the time to attend all meet-ings; (2) specialists in certain aspects of the business whowill be invited to give expert opinion in certain meetings;


(3) managers and supervisory staff who need to keepthemselves informed of the progress of the project; and (4)people outside the organization who have an interest, e.g.clients and partners.

The core team meets regularly. In the meetings, theteam analyzes and discusses the characteristics of businessprocesses; documents and edits records of discussion (i.e.the Business Functional Hierarchy and the FunctionDefinitions); as well as identifies information products anddata sources (which results in the documentation of theFunction-Information Product Table and the InformationProduct-Data Table). At certain milestones, documentsresulting from the core team meetings were sent out toother interested parties for comment. The core team isresponsible for the review and consolidation of these com-ments and suggestions into the original JAD documenta-tion.

The JAD methodology is a very effective way of per-forming business functional analysis. Since it is based onface-to-face and group meetings, it provides a workingenvironment for open communication. It also provides astructure for consensus building by focusing on issues and

resolving them. Participation in the meetings is a veryuseful educational experience for existing and potentialusers of the data to be delivered. Finally, as the deliver-ables of the core team are well documented, it provides asolid foundation for data modeling in database design. Anexample of the Abitibi Functional Analysis is provided inAppendix 1.

FUNCTIONAL EVALUATION OF GIS SYS-TEMS

In response to the Ministry’s information technology(IT) initiatives and the need for improved capture, analysisand dissemination of geological data, the OGS evaluatedthree GIS software packages. Under the supervision ofsenior management, OGS GIS Geoscientists developedevaluation criteria and bench marked the software. TheGIS Geoscientists were mandated to propose a GIS solu-tion for the OGS within the IT framework of the Ministryof Northern Development and Mines as well as the broad-er public sector.

FROM FUNCTIONAL ANALYSIS TO CD—DIGITAL COMPILATION OF THE TIMMONS MAP SHEET 115

Figure 1. Functional Hierarchy.


Function # Function Information Products

1.1.11.1.21.1.31.1.41.2.1.11.2.1.21.2.2.11.2.2.21.2.2.4.21.2.2.4.31.2.2.5.21.3.11.3.31.4.1.21.4.1.3

1.4.1.41.4.2.1.1

1.4.2.1.21.4.2.1.3

1.4.2.2

Perform publications searchSearch unpublished materialsCommunicate with peersDocument personal experienceConfirm characteristics of dataPrepare orderPerform background checkIdentify critical geologic sitesMake field notesCollect samplesSubmit samples for analysisPrepare project checklistCheck data against checklistSelect appropriate featuresMerge individual map sheets/tiles toform seamless map baseWeed map baseSpecify technique

Specify mediaConvert data sets to workingformats/mediaIdentify features of interest

list of publicationslist of unpublished materialsnotesnoteslist of data characteristicspurchasing orderlist of maps and reportslist of sitesnotes and sketcheslist of samples with descriptionGeochemical Analysis RequestData Screening ChecklistPopulated checklistlist of featuresseamless map base

weeded digital map baselist of techniques used to re-formatdatalist of media used to re-format datadata sets in working format/media

list of features of interest

Table 1. Function/Information Products Table. (Note: Table reduced in size in the interests of brevity)

Feature Type Classification Symboloutcrop areasoutcrop pointsinterpreted areasnon-interpreted areaslake areariver areariver line

Table 2. Example of Information-Product Data Table. (Note: Table reduced in size in the interests of brevity)

Methodology

In consultation with the Ministry’s Data ServicesSection, four key objectives were defined and are listedbelow:

- Measure how well the software satisfies currentgeological analysis requirements,

- Determine what existing systems could bereplaced,

- Determine what new analytical features the systemcould provide,

- Test how well it fits into our existing environment.

The following approach was developed to ensure thatthe GIS system meets these objectives:

1. Organize the evaluation criteria by the major functionsand the critical success factors the system must sup-port. The following high level functions weredefined:

- Data capture and recording of interpretations, e.g.,recording geological field observations

- Data import, e.g., bringing in satellite images orCAD drawings

- Geological analysis, e.g., creating topologicallycorrect rock unit polygons

- Data Modeling, e.g., statistical applications

- Data output and export, such as plotting on a spe-cific plotter, and exporting to Microstation

- Works within our current and future technical stan-dards

The following “soft” evaluation criteria were also con-sidered:

- Compatibility with other organizations

- Government standards

- Support, including third party support

- Training availability, including CommunityCollege and University training

- Is the data structure published?

- The company’s market share, and corporate stabil-ity

2. Pre-define the specific sub-processes or data formatsthat must be supported for each function above. Thendefine the things you want to measure subjectivelyand give them priority weighting. For example, underoutput data, ‘it must output a format usable byPublication Services Section’ and ‘it should be easy toimport into Intergraph’ (with a weight of 8 out of 10).

3. Establish a test and score the applications’ ability toaddress each of these items listed. The degree of rigorof the test normally depends on how important eachitem is and how much time is available. Existing sys-tems could be measured against these same criteria todetermine whether they could/should be replaced bythis new system.

An example of the Evaluation Form is provided inFigure 3. Those interested in a complete copy maycontact the authors for the form in a Microsoft Excelformat. For each function there is a minimumrequired score. Raster capabilities were consideredless important for our applications than vector, andconsequently they contribute less to the overall score.

4. In addition to doing an evaluation of how well eachproduct meets known needs, it was also agreed thatample time would be allowed to investigate featuresthat the tool provides beyond the immediate knownneeds. For example, one vendor demonstrated use ofthe Internet in providing GIS services, a useful capa-bility which we did not have on our list of knownneeds.

Most of the selection criteria came from the GISGeoscientists’ experience in mapping, but the follow-ing references: Bonham-Carter (1994), Eastman(1995), and Yeung (1995) quite strongly influencedGIS concepts and their application in geology. Martin(1990) also discusses critical success factor analysis,which was also used in the decision making.


Figure 2. Joint Application Design.

Performing the Evaluations

Representatives of three GIS vendors visited our sitein the fall of 1996 and winter of 1997 to help us evaluatetheir products. Two days were allotted for each evalua-tion. All vendors received our evaluation procedure, andfor those who wanted it, our test data ahead of time. Thefirst morning of each evaluation, the vendors did a promo-tional demonstration of their products’ capabilities. A typi-cal demonstration was attended by ten or twelve geolo-gists, as well as by the GIS Geoscientists. Over the nextday and a half, each vendor team performed the operationson our data as requested in the evaluation procedure.

There was a period of follow-up for several weeksafter each evaluation. For example, vendors sent us plotfiles and export files, which they did not have the time tocreate at our site. They also provided the necessary infor-mation to complete our “soft” evaluation criteria (e.g.support, training, and a list of organizations that use thespecific software).

The products showed strengths and weaknesses in dif-ferent areas of our evaluation, but tabulating the scoring ofthe functional evaluation criteria indicated which productswould best suit us.

We also considered the previously discussed “soft”factors. We wanted to be compatible with as many of ourgeological colleagues as possible and therefore consideredthe software being used by the Geological Survey ofCanada as well as neighboring provincial and state geolog-

ical surveys. Finally, we considered Ontario governmentsoftware standards and the software used by the OntarioMinistry of Natural Resources, our source for digital topo-graphic data as well as the Ontario government’s lead min-istry in GIS Information Technology.

Results

As a result of the evaluations we selected ESRIArc/Info and ArcView products as the workstation anddesktop GIS staples for the Ontario Geological Survey.This combination of products offers the best solution toour current needs as well as our requirements in the nearfuture. This solution offers a minimum impact on our cur-rent cartographic processes thus allowing us the ability tocontinue to generate hardcopy cartographic products utiliz-ing a methodology that we have developed with time,while allowing us the ability to migrate to a new deliverysystem in a functional fashion.

COMPILATION AND PRODUCTION

Compilation

The process of compilation that was used came direct-ly from the Functional Analysis (Appendix 1). The DataScreening Checklist (Table 3), also a product of the func-tional analysis, clearly lays out the data sets we had to


Figure 3. Example of Evaluation Form

work with and what features were pertinent to the compi-lation process. For our purposes, we subdivided the datainto two sets - Map Data and Tabular Data.

Map Data include both vector and raster data setsincluding the following themes:

- Topographic Map Data

- Geological Map Data

- Geophysical Map Data (aeromagnetic grids andelectromagnetic anomalies)

- Remote Sensing Map Data (Landsat TM andRadarsat)

Tabular data consisted of the following databases:

- Lithogeochemistry (ERLIS – LGC)

- Assessment File Research Inventory (ERLIS –AFRI)

- Ontario Drill Hole Database (ERLIS – ODHDB)

- Mineral Deposit Inventory Database (ERLIS – MDI2)


Data Source Scale Projection Datum DTM's

AFRIAirborne MagneticsAirborneElectromagneticsDrill Hole DatabaseExternal GeologicalInformationField ObservationsGamma RaySpectrometryGeochronologyGeology: 250000seamless baseGeology: 2000 seriesGeology: P-mapsGravityLake SedimentGeochemistryLandsat TMLithogeochemicalDatabase/LIMSMineral DepositInventoryNTSOBM (DTDB)Quaternary Geology1:50 000 seriesRadar DataTill Geochemistry

Table 3. Example of a Data Screening Checklist. (Note: Table reduced in size in the interests of brevity)

Production

Implementing the use of the new GIS software solu-tion was not immediate. The OGS continued to use theirexisting software applications such as MapInfo, AutoCADand Microstation while the skills for Arc/Info and Arcviewwere developed. By the time the second of four sheets hadbegun, these skills were sufficient that a considerableamount of the work was done solely in Arc/Info andArcview. This has facilitated the release of the GIS prod-uct yet has had minimum impact on the established carto-graphic process, a process that continues to produce fullcoloured hard copy maps for which no GIS product isrequired. Eventually, with experience gained from ournewly implemented tools, all hard copy maps will have acorresponding digital GIS product.

CONCLUSIONS

The outcome of this project is a scientific document;i.e. a geological map. The procedures we followed how-ever, in going from an analogue world of paper maps atthe beginning to a digital, attributed map at the end reallyinvolved a major paradigm shift in the way “business” ormap production at the Ontario Geological Survey occurs.This paper was written to capture the process by whichthis change occurred. It is neither meant to be a templatefor describing how such changes should take place nor is itheld up to be necessarily the correct way to effect suchchanges.

ACKNOWLEDGMENTS

We would like to acknowledge our managers A. Fyon,Senior Manager and P. Thurston , Supervising Geologist ofthe Precambrian Geoscience Section and C. Baker, SeniorManager, Sedimentary Geoscience Section for their sup-port and for allowing us to spend a significant amount oftime, first in doing the Functional Analysis and second inwriting this paper. Thanks are extended to F. Merlino,Senior Manager , Data Services Section for championingthe Functional Analysis procedure and providing staff timeand input.

NOTE: As stated throughout, numerous tables andfigures have been reduced in size and breadth in the inter-ests of brevity. Should you require comprehensive docu-mentation, tables will be provided in MS-Excel and MS-Word by simply contacting one of the authors.

REFERENCES

Bonham-Carter, G.F.,1994, Geographic Information systems forgeoscientist: modeling with GIS: Pergamon publications,

Elsevier Science Limited, The Boulevard, Langford Lane,Kidlington, 0X5 1GB, United Kingdom.

Brodaric, B., 1997, Field data capture and manipulation usingGSC Fieldlog v3.0, in D.R. Soller, editor, Digital MappingTechniques ’97, Proceedings of a Workshop on DigitalMapping Techniques: Methods for Geologic Map DataCapture, Management and Publication: U.S. GeologicalSurvey Open-File Report 97-269, p. 77-81,<http://ncgmp.usgs.gov/pubs/of97-269/brodaric.html>.

Eastman, J.R., 1995, Idrisi for Windows User’s Guide Version1.0: Clark Labs for Technology and Geographic Analysis,Clark University, 950 Main Street, Worcester,Massachusetts 01610-1477, United States.

Eckstrand, O.R., Sinclair, W.D. and Thorpe, R.I., 1995, Geologyof Canadian Mineral Deposit Types: Geological Survey ofCanada, Geology of Canada, no. 8.

Gupta, V.K., 1996, Ontario airborne magnetic and electromagnet-ic surveys: Archean and Proterozoic “Greenstone” belts; inSummary of Field Work and Other Activities 1996: OntarioGeological Survey, Miscellaneous Paper 166, p. 168-176.

Gupta, V.K., 1995, Processing of airborne magnetic and electro-magnetic surveys-progress report; in Summary of FieldWork and Other Activities 1995: Ontario Geological Survey,Miscellaneous Paper 164, p. 299-300.

Jackson, S.L., and Fyon, J.A., 1991,The western AbitibiSubprovince in Ontario; in Geology of Ontario: OntarioGeological Survey, Special Volume 4, Part 1, p. 404-482.

Jackson, S.L., Muir, T.L. and Romkey, S.W., 1995, A library ofdigital mapping symbols. Part I: digital files, figures anddescriptions: Ontario Geological Survey, Open File Report5909, 56 p.

Madon, Z., Trowell, N.F. and Ayer, J.A., 1997, Using Radarsatdata for geoscientific applications, Abitibi greenstone belt,Ontario; p. 10 –13 in Summary of Field Work and OtherActivities: Ontario Geological Survey, Miscellaneous paper168, 149 p.

Martin, J, 1990, Information Engineering, Book II: Planning andAnalysis: Prentice-Hall, Incorporated, College andTechnical Reference Division, Englewood Cliffs, NewJersey 07632, United States.

Muir, T.L., 1995, A library of digital mapping symbols. Part 2:Appendix B (listing and explanation of symbol acronyms):Ontario Geological Survey, Open File Report 5910, 116 p.

Rogers, M.C., Thurston, P.C., Fyon, J.A., Kelly, R.I. and Breaks,F.W., 1996, Mineral deposit models of metallic and industri-al deposit types and related mineral potential assessmentcriteria: Ontario Geological Survey Open File Report 5916.

Thurston, P.C., 1996, The Abitibi Field Investigations: aRationale; in Summary of Field Work and Other Activities:Ontario Geological Survey, Miscellaneous Paper 166, p. 3-4.

Yeung, A., 1995, Introduction to Geographic InformationSystems, Instructional Materials and Course Notes forfourth-year Geography course: Department of Geography,Laurentian University, Sudbury, Ontario.


APPENDIX 1. ABITIBI FUNCTIONALANALYSIS

Abitibi Compilation - Project Documentation

(Note: the content of the following has been greatlyreduced in the interests of brevity. Should you requireadditional detail, please contact one of the authors at theOntario Geological Survey)

A. GENERAL INFORMATION1. Business Activities2. Goals3. Objectives and Scope4. Usefulness5. Clients6. Custodians7. Sponsor

B. FUNCTIONAL HIERARCHY8. CREATE A GEOSCIENTIFIC DATABASE8.1 Identify sources of data8.2 Acquire data

8.3 Screen data8.4 Prepare data8.5 Compose final digital compilation maps

9. REVISE LITHOSTRATIGRAPHIC/TECTONICINTERPRETATIONS9.1 Identify new business needs9.2 Modify project objectives/specifications as needed9.3 Identify stratigraphies and tectonic environments inexisting interpretations9.4 Create working lithostratigraphic interpretation fromgeological compilation9.5 Produce digital information products9.6 Refine contents of working lithostratigraphic interpre-tation

10. CREATE METALLOGENIC INFORMATIONPRODUCTS(The analysis of this function is in progress. Function def-initions and tables to be completed)10.1 Identify new business needs10.2 Modify project objectives/specifications as needed10.3 Use computer-assisted technologies10.4 Produce information products


The Kentucky Geological Survey’s primary geologicmapping database will be connected to the KGS compre-hensive geologic/hydrologic relational database. Thiscomprehensive database (Fig. 1) has evolved considerablyduring the 20 years KGS has been computerizing data.Currently it contains data from more than 200,000 geo-graphic locations throughout the Commonwealth. Thesedata consist of well locations (petroleum and water), wellconstruction information, coal-thickness measurements,sample descriptions, analyses (coal, water, petroleum), anddescriptive information. Plans for the near future are toinstall a spatial data engine (SDE) that will provide thecapability to store spatial information in the relationaldatabase. In addition, plans are to link images derivedfrom scanning official paper documents (for example,plats, drillers logs, well tickets, down-hole logs) to theirrespective computerized records in the database. This willprovide users with immediate access to copies of the origi-nal records as received by KGS.

Retrieving and manipulating data from the main rela-tional database is accomplished by client applicationsusing ODBC (open data base connectivity) drivers, SQL(structured query language) code, or programming lan-guages such as Visual Basic (VB) or C++.

Geologic map information is collected as vector databy a GIS and stored in the KGS digital geologic mappingdatabase. These spatial data consists of numerous cover-ages such as formations, structures, faults, coals, minerals,fossils, drill holes, and dikes. Each coverage or layer con-sists of polygons, arcs, or points, and has attribute tablesthat describe the coverage. The formation code(FMCODE) and geologic quadrangle map name(GQNUM) are primary attribute headings and serve to linkother descriptive tables such as type, name, and style.

These attribute tables are the basic components of themapping database, and are directly related by scripts or hotlinks to additional digital mapping subdatabases containinginformation on subjects such as metadata, lithology,stratigraphy, mineral veins, and fossils (Fig. 2).

The Stratigraphic table and code of the spatial data-base is very important in that it is a primary table that willlink the data captured in the KGS spatial mapping data-base with the U.S.Geological Survey (USGS) draft datamodel (Fig. 3).

123

Integration of Relational Geologic Databases and aSpatial Map Database in Kentucky

By Warren H. Anderson, Thomas N. Sparks, Jason A. Patton, Xin-Yue Yang, andRichard E. Sergeant

Kentucky Geological SurveyRoom 228 Mining and Minerals Resources Building

Lexington, KY 40506Telephone: (606) 257-5500


Figure 1. Organizational Structure and access pointsfor the KGS Comprehensive and GIS MappingDatabases.


Figure 2. Components of Kentucky Geological Survey Spatial Database forGeologic Map Information.

Figure 3. The relationships between the KGS Digital Geologic Mapping Database Structure (formation,structure, faults, dikes and coal arcs and polygons) and three primary components of the USGS draft datamodel (lithology, rock unit, and strata age). These three tables serve as the primary link to access auxil-lary tables in the KGS spatial and relational databases.

Other subdatabases are designed to meet variousneeds; for example, the metadata files use a State-suggest-ed software for metadata compilation. The lithology data-base is created to supplement the digital mapping files, andthe fossil database is linked to a USGS fossil identificationnumber and description.

A GIS can be used to perform spatial queries, to locateoil-or water- well information and coal-bed thickness ormineral data; and these data are linked to other geologic,engineering, and mineral databases for resource manipula-tions, searches and modeling. Pseudo 3-dimensional mod-els have been created using digital elevation models and 2-dimensional digital geologic maps to give the appearance ofa three-dimensional view. This allows planners, developers,miners, and engineers to conduct slope stability studies, approxi-mate cut and fill calculations, and use the GIS analytical toolsto manipulate geologic data. Digital elevation models andgeologic data have been manipulated to obtain line-of-sightsurface profiles and area measurements. Plans for adding z-values will enable volume measurements to be performed.

Currently, the KGS can integrate its spatial mappingdata files with the Kentucky Department ofTransportation’s (DOT) engineering database so that rock-fall and landslide parameters can be merged into our digi-tal geologic mapping files. DOT is currently evaluatingthese data to determine how they can be used to increasethe efficiency of highway planning and maintenance.These geologic map files have also been used as a base tointegrate other nongeologic coverages such as environ-mental wetlands, parcel mapping, cultural, and best costestimates for planning highway corridors.

Customized scripts have been created to develop atemplate for automatic formatting of hard-copy output ofdigital map data. These scripts will enable a user to speci-fy the name of the quadrangle, author, and date; modify ageneric legend and stratigraphic information so that a draftquality quadrangle map can be plotted (Fig. 4). This draftmap does not have a geologic column or cross section, butwe are working on methods to create these two compo-nents.

INTEGRATION OF RELATIONAL GEOLOGIC DATABASES AND A SPATIAL MAP DATABASE IN KENTUCKY 125


Figure 4. Preliminary draft of a geologic map using an Avenue script to create a template for rapid plotting of dif-ferent geologic maps.

127

INTRODUCTION

Recent advances in handheld and Personal DigitalAssistance (PDA) computers have revolutionized the waypeople work, communicate, and store and collect data.That revolution is making its way into the business ofmaking geologic maps. During the summer 1998 fieldseason, the U.S. Geological Survey (USGS) utilized PDAcomputers in conjunction with Global Positioning System(GPS) receivers to conduct 1:24,000-scale bedrock geolog-ic mapping in southern and central New Hampshire. Theeffort was considered experimental at first, but we rapidlylearned that the combined use of a PDA computer andGPS receiver was an efficient way to collect geologic fielddata.

Field data collection systems have been employed ingeologic mapping for several years. Brodaric (1997)described a comprehensive software package calledFieldlog, developed and used by the Geological Survey ofCanada. Fieldlog utilizes an Apple Newton Message Padcomputer running Fieldworker data collection software.Unfortunately for users of this hardware and software, andfor those in the stages of planning an upcoming field sea-son, Apple discontinued the production and support of theNewton in the spring of 1998 so we turned to the burgeon-ing market of PDA computers. Here we describe the

results of our field use of a PDA computer and a GPSreceiver for the collection of geologic structure data.

The primary goal of utilizing a GPS receiver and afield data collection system was to address the problem ofdigitally compiling, in a time efficient manner, numerousgeologic structure measurements taken during the courseof mapping. The compilation of structure data is often themost time-consuming process in the production of digitalgeologic maps, especially in areas underlain by complexlydeformed and metamorphosed rocks with multiple histo-ries of ductile and brittle deformation. In order to elimi-nate the need to either spend additional field time enteringdata on a laptop computer or performing heads-up digitiza-tion of drafted and scanned structure symbols in the office,we decided to collect structure data real-time in the field.In order to achieve this goal we needed a highly portablesystem that would allow, rapid collection of geologicattribute data and positional point data. A secondary goalwas to use a data collection system that used popular hard-ware and software, making it easy for field geologists tolearn and customize it to meet their needs.

HARDWARE AND SOFTWARE

The data collection of point attributes for geologicstructures was designed around the need to create an

Geologic Mapping and Collection ofGeologic Structure Data with a GPS Receiver and a

Personal Digital Assistance (PDA) Computer

By Gregory J. Walsh1, James E. Reddy2, and Thomas R. Armstrong2

1U.S. Geological SurveyP.O. Box 628

Montpelier, VT 05601Telephone: (802) 828-4528


2U.S. Geological SurveyNational Center MS 926A

Reston, VA 20192e-mail: [email protected]

[email protected]

Arc/Info point coverage of structural geology. Althoughuntested, the system would probably work well with otherGIS packages. The PDA computer of choice was the3Com Palm III running Pendragon Forms data collectionsoftware version 1.2. Pendragon Forms is a form-basedsoftware package that can utilize the database functionalityof Microsoft Access in Windows 95, 98 or NT. Data col-lection forms that include lookup or pulldown lists andnumeric key pads can be created in Pendragon Forms,Microsoft Excel, or any ASCII text editor and then trans-ferred to the Palm Pilot. This allows users the flexibilityto use the software or the particular data fields that theyprefer and are most familiar with. Positional data was col-lected with Rockwell PLGR+96 GPS receivers using thePrecise Positioning Service (PPS) available to workers inthe U.S. Federal government. The use of the PPS allowedcollection of GPS data with a real-time advertised accura-cy of less than 20 meters, and an experienced accuracy ofgenerally less than 5 meters. This accuracy enabled map-ping at 1:24,000 without the need for post-processing theGPS data. The data collection system we employed wouldstill work with GPS data collected under SelectiveAvailability (SA), but would require the additional stepsnecessary for post-processing the data if mapping was con-ducted at 1:24,000 or larger scales. Waypoints recorded atoutcrops where structure data were gathered were electron-ically transferred, or dumped, from the PLGR+96 to a lap-top computer in ASCII format using MicrosoftHyperterminal.

PROCEDURES

The initial phase of developing a data collection sys-tem on the Palm III involved creating a form in PendragonForms. The data collection form was designed for geolog-ic structure data that would be compatible with an Arc/Infopoint attribute table, or PAT. The items in the PAT, accord-ing to the data model we developed for mapping in NewEngland’s complexly deformed rocks, are shown in Table1. This data model is generally compatible with theStructural Detail Table in the proposed digital geologicmap data model of Johnson and others (1998), but containsmore items pertaining to GPS positions and detailed struc-tural analysis in complexly deformed rocks. In creatingthe data collection system in Pendragon Forms, the userhas options to generate lookup lists, numeric keypads,popup lists, and many other types of common data entryforms. These forms, lists, and keypads allow the user tominimize the amount of handwriting or typing, and thusexpedite data entry. Figure 1 shows a partial example ofthe lookup list for the geologic structure item SUB_TYPE.Because the forms are so easy to create, the availableoptions in the lookup lists can be modified to fit any par-ticular data model.

Once the forms are complete and they match yourdata model, you can collect your attribute data and posi-tional waypoint data in the field at every outcrop with thePDA computer and the GPS receiver. The PDA and theGPS need not be physically connected in the field, unlikemany data loggers that come as accessories to GPSreceivers. This is significant, as it eliminated the need fora connecting wire between the PDA and the GPS, andallowed greater freedom of movement through rugged ter-rain and dense undergrowth.

After data collection, the PDA and the GPS data aredownloaded to a desktop or laptop computer runningMicrosoft Access and Microsoft Hyperterminal, respec-tively. Downloads could be completed as often as everyday, or as infrequently as once a week depending onaccess to a computer. We found that access to a laptop inthe field office was the most reliable way to ensure ade-quate backups of the data. Data from the Palm III down-loads directly into a Microsoft Access database file, where-as data from the PLGR downloads to an ASCII text file.Positional data from the GPS receiver represent a singlepoint and the attribute data may represent one or manygeologic measurements at that point. The attribute data inAccess and the positional data in ASCII format are thencombined into a single database by establishing a one-to-many relationship in Access using the station and way-point label as the relate item. After field work is complet-ed, or at any time when a compilation map is desired, theMicrosoft Access data can be converted into an Arc/Infopoint coverage.


Figure 1. Partial example of a lookup list for the geolog-ic structure item SUB_TYPE. The SUB_TYPE lookuplist includes valid data entry values for planar data such asbedding and schistosity shown above, linear data such asfold axes and intersection lineations, and other data suchas mines and quarries to name a few.

The first step in creating the Arc/Info point coverageis to create two separate files from the Access database: 1)an ASCII text file of positional data, and 2) a DBASE IVfile of attribute data. This is accomplished by running twoselect queries in Access and then exporting the two result-ing tables to new ASCII and DBASE IV files. In our pro-cedure we name the ASCII file coor.txt and the DBASE IVfile attr.dbf. In New Hampshire, we collected GPS data inUTM coordinates so our coordinate file (coor.txt) has threecomma-delimited fields: station, east, north. Beforeexporting, ensure that both files, or select query tables,have the same number of lines and have been sorted by thestation field in ascending order. Next add the word “end”as the last line of the coordinate select query table andexport the file; the resulting coor.txt file should look likethis:

4001,308764,4743207 — first line 4001,308764,47432074002,308649,47433844002,308649,4743384...……6534,312728,4737407

6534,312728,47374076534,312750,4737419end — last line

The format (field size or width and data type, such asnumeric or text) of the attr.dbf fields does not have to beexplicitly defined in Access or in DBASE because it willbe defined when the DBASE file is converted to an INFOfile in Arc/Info.

The generalized procedures for collecting and trans-ferring data to an Arc/Info point coverage are outlined inFigure 2. The generation of the point coverage can beautomated by running an Arc/Info macro or AML. TheAML (convertdbf.aml) uses the Arc/Info commands DBASEINFO, GENERATE, BUILD, and JOINITEM andis illustrated below:

/***********************************/* Name: convertdbf.aml/* Purpose: Converts Palm III PDA and PLGR /* GPS Access data into an ARC point cover/* Requires the following files:/* attr.dbf {DBASE IV format}/* coor.txt {ASCII format}/* Usage: &r convertdbf <outcover>

GEOLOGIC MAPPING AND COLLECTION OF GEOLOGIC STRUCTURE DATA WITH GPS AND A PDA

Table 1. Items used in the data collection system and the Arc/Info point attribute table (PAT). The descrip-tion refers to the positional or geologic nature of the item. The source refers to the source of the data eitherfrom the GPS waypoint or the Palm III data entry form.

ITEM: DESCRIPTION: SOURCE:STATION station identifier numeric keypadPOSFMT GPS position format waypointZONE GPS utm zone waypointHDATUM GPS horizontal datum waypointHRZERR GPS horizontal error waypointELEV GPS elevation waypointVDATUM GPS vertical datum waypointELEVUNIT GPS elevation units waypointTYPE geol. structure (planar, linear, other) lookup listSUB_TYPE geol. structure (joint, bedding, etc.) lookup listSTRIKE geol. structure (0-359 in right-hand rule) numeric keypadDIP geol. structure (0-90) numeric keypadDIPDIR geol. structure (0-359 in right-hand rule) numeric keypadREL_AGE relative age of geol. structure lookup listSYMBOL symbol number from Arc/Info markerset numeric keypadSYMBOL_ANG Arc/Info symbol rotation value calculated in Arc/InfoROTATION relative rotation of geol. structure lookup listSPACING spacing of joint sets or fracture zones numeric keypadWIDTH width of joint sets or fracture zones numeric keypadNETSLIP net slip of small-scale fault numeric keypadAPERTURE aperture of joints or fractures numeric keypadMIN1 primary mineral in vein or fracture lookup listMIN2 secondary mineral in vein or fracture lookup listMIN3 tertiary mineral in vein or fracture lookup listCODE relate item to geologic polygons freehand text

129


/* Project: USGS New Hampshire/* Date: September, 1998 /* Authors: Jim Reddy and Greg Walsh/* /***********************************&args cover/* OPEN TABLES - CHECK FOR /* EXISTING ATTR.DBF DATA TABLEtables info&if [exists attr -info] &then kill attrquit

/* CONVERT THE ATTR.DBF FILE /* TO AN INFO DATA TABLE/* Modify the following definitions to match /* your ATTR.DBF file and your data modeldbaseinfo attr.dbf attr definestation station 6 6 n 1posfmt posfmt 8 8 czone zone 4 4 chdatum hdatum 8 8 chrzerr hrzerr 4 4 ielev elev 5 5 ivdatum vdatum 8 8 celevunit elevunit 2 2 c type type 10 10 csub_type sub_type 35 35 cstrike strike 3 3 idip dip 2 2 idipdir dipdir 3 3 irel_age rel_age 10 10 csymbol symbol 3 3 isymbol_ang symbol_ang 3 3 irotation rotation 3 3 cspacing spacing 5 5 cwidth width 5 5 cnetslip netslip 5 5 caperture aperture 5 5 cmin1 min1 3 3 cmin2 min2 3 3 cmin3 min3 3 3 ccode code 8 8 cend

/* GENERATE THE POINT COVERAGE /* FROM THE COOR.TXT INPUT FILE&if [exists %cover% -cover] &then kill %cover% allgenerate %cover%input coor.txtpointsquit

/* BUILD THE POINT COVERAGEbuild %cover% point

/* JOIN THE POINT COVERAGE PAT FILE /* AND THE ATTR DATA TABLEjoinitem %cover%.pat attr %cover%.pat %cover%# %cover%-id link

&return

Once the point coverage is created, SYMBOL valuescan be checked or added based on values from markersymbols in an Arc/Info markerset of geologic symbolssuch as those created by Fitzgibbon and Wentworth(1991). Before plotting the symbols in the correct orienta-tion, the item SYMBOL_ANG must first be calculatedfrom STRIKE values for planar symbols and DIPDIR val-ues for linear symbols. This calculation depends largelyon the orientation of the symbols in the markerset and themethod of collecting strike data. After calculation of theitem SYMBOL_ANG, the pseudo item $ANGLE can becalculated as equal to SYMBOL_ANG. The pseudo item$ANGLE records the rotational angle for individual sym-bols in Arc/Info and is based on a Cartesian coordinatesystem with East = 0°, North = 90°, West = 180°, andSouth = 270°. This differs from standard geologic angleswhere North = 0°, East = 90°, South = 180°, and West =270°.

RESULTS

At the time of this report, mapping in two 7.5-minutequadrangles (Windham and Pinardville) in southern New

Figure 2. Flow chart showing the generalized proce-dures for collecting and transferring data to an Arc/Infopoint coverage.

Hampshire and the Hubbard Brook watershed in centralNew Hampshire has successfully employed this field datacollection method. In the case of the Windham quadran-gle, mapping was completed in September 1998 and acomplete digital geologic map with over 2400 structuresymbols was compiled and submitted for review byFebruary 1999 (Walsh and Clark, 1999). Digital compila-tion of the Pinardville quadrangle is almost complete andadditional geologic mapping in the Hubbard Brook water-shed is scheduled for the 1999 field season.

In the past, data collection typically involved record-ing structure data in field notebooks by hand followed bydata entry into a computer database or heads-up digitizingof drafted and scanned structure symbols after the comple-tion of field work. The old method often led to lengthyperiods of data processing and even simplification of thedataset in order to meet budgetary constraints and dead-lines. The Hartland quadrangle in Vermont is underlain bysimilarly complex geology and the Arc/Info point coveragecontains 1600 points (Walsh, 1998). Compilation of thestructural geology point coverage took a total of approxi-mately 40 hours and included hand-drafting of symbolswith pen and ink on mylar (24 hours), scanning and regis-tering the drafted symbols (1 hour), and heads-up digitiz-ing (15 hours). For comparison, compilation of the pointcoverage for the Windham, New Hampshire quadranglerequired only approximately 10 hours and included period-ically downloading GPS and Palm III data (8 hours), datamanipulation in Microsoft Access (1 hour), and generationof the Arc/Info point coverage (1 hour). In this compari-son, the old method used in Vermont took four times aslong to create a point coverage with only two-thirds thenumber of points.

CONCLUSIONS

The USGS developed a field data collection systemfor geologic mapping that utilizes GPS receivers and PDA

computers. The system uses a PLGR+96 GPS receiverwith the Precise Position Service and a 3Com Palm IIIPDA computer running Pendragon Forms data collectionsoftware. Data transfer to an Arc/Info point coverage ofstructural geology is accomplished by data manipulation inMicrosoft Access, data extraction to ASCII text andDBASE IV files, and data conversion using an Arc/Infomacro (AML). The method has allowed for the rapid,accurate collection of abundant structural geology data incomplexly deformed and metamorphosed rocks. Thismethod greatly expedites the digital compilation of geo-logic map data in a GIS and, although untested in othergeologic settings, should be easy to apply to other areas.

REFERENCES

Brodaric, Boyan, 1997, Field data capture and manipulationusing GSC Fieldog v3.0, in D.R. Soller, editor, DigitalMapping Techniques ’97, Proceedings of a Workshop onDigital Mapping Techniques: Methods for Geologic MapData Capture, Management and Publication: U.S.Geological Survey Open-File Report 97-269, p. 77-81.

Fitzgibbon, T.T., and Wentworth, C.M., 1991, ALACARTE UserInterface: AML code and demonstration maps, U.S.Geological Survey Open-File Report 91-587.

Johnson, B.R., Brodaric, Boyan, and Raines, G.L., 1998, Draftdigital geologic map data model, versions 4.2 and 4.3:Unpublished American Association of State Geologists /U. S. Geological Survey draft document, 83 p.,<http://ncgmp.usgs.gov/ngmdbproject>

Walsh, G.J., 1998, Digital bedrock geologic map of the Vermontpart of the Hartland quadrangle, Windsor County, Vermont:U.S. Geological Survey Open-File Report 98-123, 17 p.,scale 1:24,000.

Walsh, G.J., and Clark, S.F., Jr., 1999, Bedrock geologic map ofthe Windham quadrangle, Rockingham and HillsboroughCounties, New Hampshire: U.S. Geological Survey Open-File Report 99-8, scale 1:24,000.

GEOLOGIC MAPPING AND COLLECTION OF GEOLOGIC STRUCTURE DATA WITH GPS AND A PDA 131

133

INTRODUCTION

This is the second year that geologists at Missouri’sGeological Survey Program (MGSP) have used ArcView(version 3.0 in 1998, and version 3.1 in 1999) to producegeologic maps for the STATEMAP project. STATEMAPis a component of the National Cooperative GeologicMapping Program. Each geologist takes his or her mapfrom the initial entry of field data to the final publishedproduct. It has been found that the “heads up” mapping ona USGS Digital Raster Graphic topographic map (DRG)displayed on the monitor has many advantages over con-ventional pen and ink mapping. Geologists using thesetechniques can readily create a variety of graphically dis-played working hypotheses to improve the accuracy of theinterpreted geologic map. The ability to use separatethemes to quickly draw, view, modify, and erase interrelat-ed features is especially valuable. Geologists can easilycompare common boundaries between their map areas andassure that the interpretations match. This facilitates com-pilation of 1:24,000 sheets into composites. The tech-niques used by Missouri’s Geological Survey Programhave involved the typical trial and error of any new tech-nology. These techniques are rapidly evolving as the geol-ogists become increasingly familiar with the capabilities ofthe software and each makes contributions from his or herown area of expertise. Some of the experiments of thepast two years are discussed in the following paragraphs inanticipation that they might prove useful to geologistsmapping in other states.

FROM FIELD DATA TODIGITAL INFORMATION

MGSP geologists collect field data by drawing out-crops on a 7.5 minute USGS topographic map and identi-fying each outcrop with a number. ArcView and USGSDRG’s have made it possible to easily print sections of thetopographic maps on 11x17 inch paper at 1:12,000 for usein the field. This enlarged scale makes it easier to drawand label outcrops legibly on the field maps, especiallywhere there is a high density of outcrops. A field note-book is maintained with a description of each outcrop andgeneral traverse information.

Upon returning to the office, an outcrop map is pre-pared by copying the marked outcrops from the field mapto the DRG by drawing a polygon in ArcView with themouse (figure 1). The size of small outcrops is exaggerat-ed so that they will be visible while drawing the bedrockcontacts. Outcrops are typically entered with the viewzoomed in to a scale of 1:4000 or less. Key informationincluding ID number, formation, lithology, and structuresare entered into the attribute table. Traverses are separatedas to whether they were made by car, boat, or foot and areentered as line themes. Recording traverses on the mapdocuments where the geologist went in the search for out-crops, whether or not any were found. Complete fieldnotes are typed into a text file using Microsoft Wordpadand then copied to individual outcrop or traverse files sothey can be displayed in ArcView by merely clicking onthe feature with the “hot link” tool. Sketches, annotated

ArcView, a Geologic Mapping Tool

By Harold W. Baker

Missouri Department of Natural ResourcesDivision of Geology and Land Survey

Geological Survey ProgramP.O. Box 250

Rolla, MO 65402Telephone: (573)368-2145

Fax: (573)368-2111e-mail: [email protected]


RIV

ER

Figure 1. Enlargement of a portion of the Grandin SW 7.5 minute quadrangle showing outcrop polygons forthree map units. The four outcrops shown in black, closest to the river are Lower Gasconade Dolomite. Thedark gray outcrop in the middle is Upper Gasconade Dolomite and the light gray outcrop at the highest eleva-tion is Roubidoux Formation (the outcrops are normally displayed in easily distinguishable colors). All are ofOrdovician age. The outcrop map is the basic map of the field data and is used as the basis for all of the inter-preted maps.

photos, and measured sections can be drawn or enteredinto a separate view and linked to the outcrop to permitretrieval by a click of the mouse. These data and displaysare used in preparation of the final interpreted maps andare available for reference by other geologists working inthe mapped areas on environmental projects, mineralexploration, etc. The actual format in which these support-ing data will be made available to the public has not beendetermined at this time, but will probably be on a CD.

INTERPRETATION TECHNIQUES

The purpose of the mapping is to produce a bedrockgeology map that shows the formation (or other mappableunit) occurring at the surface as well as any mappablestructures. The bedrock geology map units are drawn aspolygon themes. Structures are drawn as line themes.

Each map unit is drawn as a separate theme startingfrom the oldest up. The lowest formation is drawn as apolygon with vertices at only the four corners of the map.Each stratigraphically higher unit is drawn to enclose theareas that lie topographically above its basal contact. Inthis technique, editing is quick and easy. The themes aremerged after editing is complete and before distributing orprinting the maps.

Mapping is straightforward in areas where outcropsare abundant, control on contact elevation is good, and thestrata are close to horizontal. The contacts can be drawnas bedrock geology polygon themes by following the DRGcontours and visually interpolating the elevation betweenthe outcrops, which are color coded for each map unit.Faults are drawn as line themes and the bedrock polygonscan be closed on either side of faults or carried across thefault with the appropriate offset depending on the map-per’s preference.

In areas where outcrops are widely scattered or sparseand the strata are folded, ArcView is a very useful tool forgraphically overlaying different data sets and workingthem against each other to find the best fit map for all thedata. Adding themes for structure and isopach contourswas very helpful on one quadrangle where mapping wascomplicated by a combination of broad folds and mapunits of variable thickness. With the structural and strati-graphic complications and the complex dendritic drainagepattern, it was difficult to simply “eyeball” the elevation ofthe contacts correctly between the moderate number ofoutcrops.

One of the contacts was exposed in several outcropsthroughout the quadrangle. By preparing a structure con-tour map (figure 2) on this horizon and overlaying thestructure map on the topography, it was easier to interpretthe elevation of the contact. The first step in creating thestructure map was to visually scan the outcrops for thosewhere a reliable elevation could be picked and enter thoseoutcrops as a separate point theme with the elevations

entered in the attribute table. Each point was then labeledwith the elevation. A structure contour map was thendrawn as a line theme, using these data points as a base.

In order to draw the bedrock geology polygons, theoutcrop theme was turned on with each of the outcropscolor coded by formation. The structure contour lineswere also color coded to facilitate recognition and makemapping easier. The DRG was placed on top and thewhite and green colors were made transparent. Thebedrock polygons were then drawn with the contact at thecorrect elevation on the DRG by visually interpolatingbetween the structure contours and the outcrops.

A similar approach was used to draw the contactwhich overlies a map unit of variable thickness. Becauseof thickness variations, the contact could not simply bedrawn a fixed vertical distance higher throughout the map.In this case, there were enough outcrops where a maxi-mum, minimum, or actual formation thickness could bedetermined. Those outcrops were entered as a separatepoint theme and labeled. The isopach maps were drawn asa line theme from these data points (figure 3). Isopachcontours were labeled but left black. The contact abovethis unit of variable thickness could then be drawn byvisually interpolating between the structure contours, theisopach contours, and the outcrops (figure 4).

ARCVIEW, A GEOLOGIC MAPPING TOOL

Figure 2. Map of the Grandin SW 7.5 minute quadrangleshowing structural control points, and structure contours.

135


Where discrepancies occurred during preparation ofthe geologic map, the attribute table and field notes werechecked first for descriptive information about the out-crops. In some cases, the outcrop observations were iden-tified as being ambiguous or questionable and the outcropinterpretation was changed to fit the structure and isopachmaps. If the outcrop data was not ambiguous, the struc-ture and/or the isopach was modified until all of the inter-preted maps fit the data. Simultaneously working withthese electronic overlays was much easier than workingwith similar overlays on film or tracing paper.

CROSS SECTIONS AND SCALEDDRAWINGS

In the past, measured sections, stratigraphic columns,and other scaled drawings have been drawn in AutoCADor by hand and then drafted in Micrografx Designer soft-ware. Redundant steps can be eliminated if geologistshave a way to make these drawings directly in the comput-

er. ArcView is much more user friendly for most of thegeologists than AutoCAD, and drawings that are shapefilesin ArcView are convenient for coloring, labeling, andusing either in views or layouts. Unfortunately, no accept-able way was found to make scaled drawings, directly inArcView. This problem was partly solved by creatinggraph paper drawings in AutoCAD and importing theminto ArcView. These graph paper templates have beenused successfully to draw scaled stratigraphic columns andmeasured sections directly in ArcView. The graph paperbackground can be turned off for display or printing.

Experiments are currently underway using AutoCADMap 3.0 to draw the scaled elements of cross sections.AutoCAD Map has the capability to rotate a DRG. Byrotating the DRG until the line of the desired cross sectionis in a horizontal position, it is possible to place a scaledgrid along the line and use the cursor cross hairs to plotthe topographic profile and position of geological features.Once the scaled aspects of the drawing are complete, thedrawing can be returned to ArcView to fill in detail, text,and color. Other software alternatives are also beingreviewed including the use of a USGS DEM in Arc/Infoand 3D Analyst ArcView Extension.

COMPILATION OF COMPOSITE MAPS

Missouri has 6 geologists assigned to map 7.5 minutequadrangles as part of the STATEMAP project. We havegeologically mapped adjoining quadrangles within a1:100,000 quadrangle area and then compiled the individ-ual 7.5 minute maps into the composite map for publica-tion. The first of these composite maps is being completedthis year. It includes 13 maps that were drawn originallyin ArcView and 19 maps that were drawn on mylar withink and then digitized using GSMAP.

The maps drawn in ArcView were the simplest tocompile. Individual geologists were responsible for assur-ing that their maps match along all of the common bound-aries. This was easily accomplished in ArcView by addingthe shape files from the adjoining quads to each map.Where discrepancies occurred, the geologists involvedresolved the problem. In most cases it was a matter of onegeologist having control where the other didn’t and themap that lacked data was modified to fit with the one withthe best control. In a few cases, the discrepancy was a dif-ference in interpretation and occasionally required thegeologists to return to the field together and come up witha mutually acceptable interpretation. In either case, thecorrected and matched shapefiles were sent to the geolo-gist making the compilation.

The data digitized in GSMAP was more difficult tohandle and much less accurate. The data was importedinto ArcView as dxf files in the form of polylines, rather

Figure 3. Map of the Grandin SW 7.5 minute quadrangleshowing thickness control points and isopach contours ofthe Upper Gasconade Dolomite. Numbers to the right ofthe control points are actual thicknesses. Numbers aboveare maximum thicknesses and numbers below are mini-mums.

ARCVIEW, A GEOLOGIC MAPPING TOOL

Figure 4. In this clip from near the center of the Grandin SW 7.5 minute quadrangle, the structure map, isopach map,outcrop map, and topographic map are combined to provide control for drawing the formation contacts for the geologicmap. The structure contours are color coded with the even 100 foot contours as red and each of the other contours (20,40, 60, and 80) as other distinct colors that can be easily seen and remembered. This facilitates interpolation between thecontours when only 1 or 2 are visible at the scale that the map is being drawn (usually between 1:4000 and 1:8000). Theheavy angular line is the boundary of the Ozark National Scenic Riverways.

137


than polygons. It was necessary for a technician to closethese lines and convert them to polygon themes before thecompiling geologist could work with them. The quality ofthe digitizing varied from excellent to poor and the geolo-gist making the compilation had to frequently refer to theoriginal inked map to determine how to edit the boundaryareas to make them match. Another problem was thatsome of the authors of these early maps are no longer withthe Geological Survey Program and could not help withresolving boundary discrepancies. The compiling geolo-gist had to make the best interpretation possible from theexisting maps.

Future compilations should be much easier with all ofthe maps drawn, edited, and matched along the boundariesby the original mapping geologists using ArcView.

LAYOUTS AND PRINTING

Procedures for preparing the layout for the final print-ed map have evolved as well as the mapping techniques.Last year, the final product was a composite of drawingsand tables prepared by drafting personnel, copied segmentsfrom text files, and graphics drawn in ArcView. Cross sec-tions and stratigraphic columns were drawn by hand andturned over to the drafting department. Drafted illustra-tions were prepared using Micrografx Designer softwareand exported to ArcView as eps files. These files werethen edited in the layout.

The geologists involved in mapping met frequently todevelop standards so that all the maps would be as similaras possible. Now that the standards have been establishedand with the new techniques for the scaled drawings, most

of the layouts are being prepared directly by the geologistsin ArcView. Composite stratigraphic columns, legends,and templates are being prepared that will allow each geol-ogist to clip those parts that are needed for his or her mapand edit them as needed. Drafting support will be mini-mal.

SUMMARY

The Geological Survey Program in Missouri hasfound that mapping by geologists directly in ArcView hasresulted in a significant improvement in map accuracy andboundary compatibility. Plotting the correct elevation ofgeological contacts depends on structure, unit thickness,topography, and outcrops. All of these features can bemapped, and then viewed and edited simultaneously inArcView, facilitating the geological interpretation.Compilation of maps is easier when each mapping geolo-gist compares his or her mapping with the shape files ofadjacent maps and resolves any boundary discrepancies.While AutoCAD is required for some scaled drawing, tem-plates imported from AutoCAD allow most geologicalwork to be done within the simpler environment ofArcView. Preparation of layouts by geologists withinArcView eliminates much of the need for drafting supportand the redundancy of penning draft drawings to be digi-tized or drafted by others. The program has proved to bevery user friendly and geologists with little computerbackground were able to pick up the techniques and suc-cessfully complete their map assignments within the firstyear.

139

INTRODUCTION

The City of Rocks National Reserve, jointly adminis-tered by the National Park Service, Idaho Department ofParks and Recreation, and local authorities, received thenational attention to designate it as a Reserve largelybecause of its position on historic emigrant routes, includ-ing the California Trail. Emigrants passing through thestunningly scenic landscape of giant rock spires and pinna-cles dubbed the area the “Silent City of Rocks” and signedtheir names on many pinnacles. As a result, geology has akey place in the founding of this Reserve, for the geologiclandforms led to the cultural resources. Geology also is akey contributor to land-resource management decisions,because the susceptibility of surface materials to erosion,deposition, and destructive processes such as landslidingrequire geological study. The surface materials also partlycontrol vegetation distribution and habitats, and may influ-ence invasion of exotic species.

We developed custom geologic databases for the Cityof Rocks so that they can be used in GIS systems withother databases. The datasets have been organized in anArcView GIS project, a commercial GeographicInformation System (GIS) package (EnvironmentalSystems Research Institute, Redlands, California), thatsimplifies the loading, display, and use of the databases.The database and ArcView project have been developedwith the idea that users are not geologists, but do haveuses for scientific geologic information. A geologic mapdatabase can be plotted on a contour map base to create astandard geologic map. In this mountainous area geologyhas a strong correlation to terrain, so the geologic map canalso be presented as a composite with shaded relief. Mapscreated from susceptibility models for many geologicprocesses in the area display areas with potentially highvulnerability to destructive processes such as erosion and

rock falls. Physical inventory databases have also beencreated, such as an inventory of pinnacles and othergranitic landforms, and an associated photographic inven-tory that documents the present state of granitic and relat-ed surficial features. The databases are in the final stagesof production, as is the development of the ArcView appli-cation. Management staff at the City of Rocks has showedstrong interest in the database, to be released later thisyear, and in the ongoing development of scientific researchand development of resource applications in the Reserve.The databases will likely be incorporated into an InternetMap Server application that may include virtual field tripsand other interpretive information.

GEOGRAPHIC AND HISTORIC SETTING

The City of Rocks, located near the town of Almo inthe Albion Mountains of south central Idaho, was estab-lished as a National Reserve in 1988 to preserve culturaland natural resources that attracted commentary in thejournals of emigrants to California over a 50-year periodin the 19th century. By far the most important part of theemigration through City of Rocks was caused by the GoldRush of 1849, which led to pioneer wagon trains for sever-al more decades. In addition, the Salt Lake Alternate Trailpassed through the southern part of the Reserve, joiningthe California Trail there in Emigrant Canyon. This trailalso was used for several decades as part of the stagecoachroute between Salt Lake City and Boise. The emigrantsnot only commented on the fantastic pinnacles but also onlush green fields and abundant water for their cattle andoxen. They covered pinnacles near the trail with names ofrocks, their own names, and other graffiti. Some of theseinscriptions remain as a reminder of the past travelers.

Digital Resource Database for Management Decisions inCity of Rocks National Reserve

By David R. Bedford and David M. Miller

U.S. Geological Survey345 Middlefield Rd MS 975

Menlo Park, CA 94025Telephone: (650) 329- 4924


[email protected]


At the City of Rocks, geology is a focal point for cul-tural and natural resources, as well as for recreation. Thecultural resources range from the original emigrant routesthrough late 19th century settlements and the subsistencelifestyles of the early American west. Emigrants took noteof the area because of the pinnacles and abundance ofwater, both being attributes of the geology that formed theAlbion Mountains. Members of early communities adapt-ed their lifestyles to the offerings of the land, whetheravailability of clay for brick making, stone for building, orwater for irrigating. Geology has a fundamental influenceon natural resources as well, since the geology controlslandforms, and these landforms influence the plant andanimal communities. Geological features are also notedfor their recreational values. The eroded granitic terrainharbors a maze of spires, which are popular amongclimbers. However, the granitic sediments of the area arenot able to withstand many land-use practices, resulting ineroding sections of trails and roads and diminishingwildlife and vegetation.

GEOLOGIC DATABASE

In addition to the geologic features that have interest-ed travelers and local communities, City of Rocks hasattracted geologists because it is a metamorphic core com-plex. The region was identified as early as 1968 as an areaof at least three phases of early-mid Cenozoic regionaldeformation events which included denudation by crustalextension accompanied by emplacement of granite(Armstrong, 1982). The topic of metamorphic core com-plexes was widely debated in the 1970’s and still attractsmany researchers to the western United States, and to thecore complexes in and around the City of Rocks.Contrasted to the mid-Cenozoic deformation and accompa-nying mountain building, research for this study suggeststhat the three main upland basins in the City of Rocks arerelatively stable. For instance, they accumulate alluviumand colluvium very slowly. However, the area is not nec-essarily stable. Processes such as soil creep, landsliding,and debris flows can impact areas that appear to be quies-cent.

The geologic database consists of an Arc/Info(Environmental Systems Research Institute, Redlands,California) format dataset that represents bedrock and sur-ficial deposits for the City of Rocks. Geology wasmapped at 1:24,000 scale and larger when necessary.Much of the recent mapping was done using GPS and in-field databases to improve accuracy and increase efficien-cy of mapping. The digital field systems used for much ofthe mapping consists of a PLGR GPS unit with an average8m locational accuracy, and a PalmPilot handheld comput-er running a database package which serves as the geolo-

gist’s notebook. Because notes and positional informationare acquired digitally and accurately, they allowed dailymapping information to be downloaded into a laptop com-puter at the end of each day and incorporated into a spatialdatabase of observations. Nightly updates to the geologicmap were performed using the daily observations andDigital Orthophoto Quadrangles in ArcView GIS, usingArcView’s on screen digitizing capabilities for shapefiles.When mapping was completed, the geologic map wasimported into an Arc/Info database, and topological errorsinherent in shapefiles were corrected. The database con-sists of a geologic units and faults layer, structural layer,and cross-sections; all are viewable in ArcView or otherGIS packages.

GEOLOGIC STABILITY MODELS

Stability models for several types of geologic process-es were developed to help land managers at the City ofRocks to understand the interactions of terrain, geologicmaterials, vegetation and climate. These models can becombined with other survey data such as vegetationassemblages and animal habitat to develop more compre-hensive models for determining management alternativesfor uses such as: grazing allotment decisions, facilities sit-ing, road and trail siting, fire management, water use,drainage diversion decisions, and developing monitoringstrategies. Input data sets for the susceptibility modelsinclude data from: 1) U.S. Geological Survey (30 meterDigital Elevation Models, Digital Orthophoto-Quadrangles, and topographic maps): 2) Soil ConservationService (Soil survey of part of Cassia County, digitized inArcView); and 3) custom data from field studies (bedrockgeology and surficial materials, and landforms).

The basic form of each model is a series of geologic,terrain, and climatic factors that are numerically modeledto create an output susceptibility map. Input factors areindividually weighted to classify each factor into an appro-priate scale (for instance, slope is continuously changingdata, but there may be threshold slope values that need tobe expressed in the models). Individually weighting eachfactor in the model also puts all the data into the samerange of values. This is important in order to numericallycompare factors: if a geologic unit is ‘late Pleistocene allu-vium’, and you want to model the slopes on that unit, youhave to convert a text attribute into a numerical attribute.Similarly, to compare erosion potential between bedrockunits and surficial units, the units must be weighted againsteach other so that surficial units will be more susceptibleto erosion. Each factor is classified and assigned numeri-cal values based on theoretical and experimental studiesfrom the literature, from knowledge of geologic processes,and from inferences based on geologic deposits and other

information at the Reserve. Once each factor is individu-ally weighted and in the same range of values, an equationcan be developed that weights each factor against the oth-ers. This equation is then evaluated on a pixel-by-pixelbasis to arrive at the final result, which is classified byinspection and calculation into categories of high, medium,and low susceptibility and displayed in map form. Figure1 illustrates this process. Although the models are essen-tially empirical, they yield numerical results that can beused to test and calculate statistical measures of suitability,should those approaches be desirable.

The models display areas with potentially high vulner-ability to destructive processes (Figure 2), particularly ifthe land is disturbed (such as by wildfire, road building,and drainage diversion). Simple GIS overlays of thesemaps provide guides to making land-use decisions, such asrouting a road. The models are created using Arc/Info’sGRID module, which performs raster-based analysis.Results of the models can be viewed as images withinArcView, or converted into a vector format to allow the

model results to be queried. The models can be improvedby collecting better quality information on soils, moredetailed slope maps, and better precipitation data.

PINNACLES AND GRANITIC LAND-FORMS DATABASE

The most notable landform features of the Reserve arethe prominent, steep-sided, smooth and rounded graniticpinnacles. The name City of Rocks refers to this assem-blage that resembles the assortment of tall and short build-ings found in a city. These rock features are as high as500 feet and are found across nearly 2000 feet in elevationfrom low basins to high ridges. Most pinnacles haveformed in the granite of the Tertiary Almo pluton. Jointsoriented nearly north-south are the most strongly devel-oped, causing many of the pinnacles to be elongate north-south. Convex upward, dome-shaped joints and nearlyhorizontal joints are believed to be responsible for the two

DIGITAL RESOURCE DATABASE FOR MANAGEMENT DECISIONS IN CITY OF ROCKS NATIONAL RESERVE

Figure 1. Flow diagram illustrating the Geologic Susceptibility Models.

141

main types of geomorphic features in the pinnacles: born-hardts and tors respectively (Cunningham, 1971).

The pinnacles and granitic landforms database repre-sents outcrops of granite features with abrupt topographicrelief on one or more sides, as distinct from exposed pedi-ment and other nearly flat granite features. The location ofpinnacles were “heads-up” digitized using ArcView GIS,and then converted into Arc/Info format. DigitalOrthophoto Quadrangles of the Reserve were used as abase for digitizing the polygonal outlines of the bases ofpinnacles within the Reserve. Stereoscopic viewing ofmultiple years of aerial photographs (1956 - 1992) aided indistinguishing granite features of raised relief fromexposed pediment and rock fall. Digitizing was done atvariable scales from 1:3,000 to 1:10,000. Field checkingwas performed in many areas during the photographic doc-umentation of granitic rock features.

A spatial inventory of location of granite rock out-crops having raised relief can be used for a variety of pur-poses, including base map applications and spatial analysisin a GIS. Base map applications include: identifying andlocating (1) rocks with names, (2) pinnacles used forclimbing, (3) rocks bearing inscriptions, and (4) rocks withunique, rare or fragile features. The database can also beused to: (1) prepare thematic maps (or pamphlets) usingprominent pinnacles as selected reference points, (2) refer-ence features or landmarks on trail maps or climbingguides, (3) prepare road maps, and (4) aid in search andrescue operations. Below we describe its use in a rock andfragile feature inventory.

This digital database can be used in ArcView, or otherGIS packages, to query for relationships with other GISdatasets for interpretation, resource management, mainte-nance, and design & development applications. Thisdataset can also serve as a base map for baseline inventoryof other resources (e.g. wildlife habitat; pack rat middenlocalities; inscription rocks; climbing rocks, etc).

INVENTORY AND PHOTOGRAPHICDATABASE OF SPECIAL FEATURES

The pinnacles database was used to initiate arock/fragile feature inventory. This database provides pho-tographic documentation of many of the landforms in thePinnacles database. The photographs document the pre-sent condition of features that appear to be fragile, includ-ing rock faces undergoing frequent climbing activity andspecial wildlife habitat situations related to the shapes oflandforms. The database contains features such aswoodrat middens, natural arches and windows, precarious-ly balanced rocks and rock shelters. The photographicdatabase contains two types of images: photos of a particu-lar feature, and sets of photos that are of vistas or panora-mas. Photos of vistas are symbolized and queryable inArcView by both the location that the photo was takenfrom, and the general direction the photo is depicting(Figure 3).

Photographs were taken using a digital camera andaerial photographs and topographic maps were used tonote the location of each photo. A short description of thefeatures seen in the photograph was incorporated into thedatabase, as well as common keywords that can be attrib-uted to many photographs to allow ease of querying thedatabase for a particular theme found in the photos. Thephotographic database is implemented in ArcView throughthe use of hotlinking: when a user clicks on a point orpoints in the photographic database, all photos associatedwith that point are automatically loaded into separate win-dows within ArcView. Further modifications to theArcView project will probably automatically load thedescription of the photo as text at the bottom of the image.Currently there are over seven hundred photos in the pho-tographic database, including repeat photography from the1970’s and again in the late 1990’s. Another research pro-ject in the Reserve was the professional photography ofnineteenth and early twentieth century rock inscriptions bythe many emigrants through the City of Rocks. This data-base, not presently in a GIS, accentuates the inscriptionsand special filtering has allowed the analysis of theinscriptions, some of which aren’t visible to the naked eye.The database of inscription photography could also beincorporated into a GIS because we have provided a spa-tial database of rocks on which these inscriptions arefound.


Figure 2. Sample susceptibility mapshowing the reserve boundary and roads inthe Reserve.

The purpose of the inventory and photographic data-base of special features is to document the location andcurrent status of features considered to be fragile in theReserve. By incorporating photography, land managerswill be able to visually inspect the current conditions ofthese features, and in the future, will be able to analyze thechanges that have occurred on these landforms. The pho-tographic inventory provides a baseline for analysis, sothat qualitative, and possibly quantitative, analysis of thedegrees of impact or recoverability to these landforms willbe possible.

CONCLUSIONS

Geologic oriented databases have been developed tohelp land managers in the City of Rocks National Reservemake more informed decisions about land use that isaffected by geologic processes. Geologic databases, ter-

rain, susceptibility to geologic processes, derivative maps,and inventories of physical resources are included in thedatabase to give access to geologic information on a man-ager’s desktop. A GIS system based on ArcView GIS hasalso been developed to promote the use of this informationby simplifying the management, display, and use of thesedatabases.

REFERENCES

Armstrong, R.L., 1982, Cordilleran metamorphic core com-plexes—From Arizona to southern Canada: AnnualReviews of Earth and Planetary Science, v. 10, p. 129-154.

Cunningham, F.C., 1971, The Silent City of Rocks, a born-hardt landscape in the Cotterrel Range, south Idaho,USA: Zeitschrift fur Geomorphologie, v. 15, p. 404-429.

DIGITAL RESOURCE DATABASE FOR MANAGEMENT DECISIONS IN CITY OF ROCKS NATIONAL RESERVE

Figure 3. Screenshot of photographic database in use.

143

145

INTRODUCTION

At the 1998 Digital Mapping Techniques Workshop,Gail Davidson of the Division of Geological &Geophysical Surveys presented some of the experiencesand problems that the Survey has encountered during thedevelopment of a digital mapping database (Davidson,1998). The result of this presentation was a flood of goodadvice from the workshop participants. Since the ’98workshop we have been putting some of those ideas towork.

WHERE WE HAVE BEEN

One suggestion offered, during the workshop last year,was the incorporation of ArcView into our mapping pro-gram. Participants from other surveys pointed out the easeof on-screen-digitizing with this software. This was seenas an attractive tool to get project and field geologistsmore involved in entering and processing their data. Theease with which existing data sets could be viewed andanalyzed also was also seen as a useful addition to our sys-tem.

Since last year we have added ArcView to our Unixsystem and several of our staff now have it installed ontheir PC’s. The result of this seems to be increased inter-est in data entry and manipulation by formerly “non-GIS”people. The ease with which the existing data sets can beaccessed and combined offers exciting possibilities. In thepast, combining old geology and new, geophysics and geo-chemistry, or remotely sensed data and geophysics, wassomething that only the “computer guys” could do on theUnix workstation. Now, the possibility exists for anygeologist to manipulate the data themselves on their ownPCs. Many informational maps for reports, and topo-graphic bases for mapping purposes have been produced.ArcView will be in the field, on laptop computers, during

at least one project in the 1999 field season. It is our hopethat this new excitement over GIS will get our people tolook at their data not only as map-specific information, butalso as data that can be used in the future for analyticalpurposes.

While this new found enthusiasm for GIS is encourag-ing because of the increased efficiency in production that itcan bring, it also carries with it a new set of concerns.Although ArcView is fairly user friendly and the basicscan be learned very quickly, proficiency takes some time.Part of being proficient in any GIS system is an awarenessof the structure of the database and how the data beinggenerated fits into that structure. Some control has to bemaintained over the way existing data is handled andwhere new data is stored and maintained. For instance,ensuring that data sets that are sitting on someone’s PCeventually make it into the agency’s server will take bothdiscipline and training. This is not an insurmountableproblem, however it is an issue that has to be addressed.Along with the overall organization of the database is theconcern over the differences in formats between Arc/Infocoverages and ArcView shapefiles. It isn’t clear to us yetwhether the difference in formats will limit what we pro-duce in ArcView. Our existing data is in Arc/Info cover-age format. Although ArcView shapefiles can be convert-ed into coverages, it doesn’t seem to be an entirelystraightforward process. Can we produce polygon shape-files and convert them to coverages without losing accura-cy? Can we take coverages and convert them to shapefilesfor editing purposes and convert them back to coveragesall the while maintaining the integrity of the data? Beforewe start generating large shapefiles for use with existingdata, we need answers to these kinds of questions.

Another way that we have gotten geologists in closercontact with their data is by upgrading our plotter technol-ogy. Until about 18 months ago all of our plotting wasdone on a Versatec plotter. This plotter was available onlythrough the Unix system and as a result plotting jobs had

Present and Future Issues of a Mapping Program

By Frank Ganley

Alaska Division of Geological & Geophysical Surveys794 University Avenue, Suite 200

Fairbanks, Alaska 99709Telephone: (907) 451-5005



to be performed by a select few people. We are in theprocess of phasing out the Versatec. We presently areusing two HP 2500’s in addition to the Versatec and byJuly 1 will be using the HPs exclusively. These two plot-ters are available through all of the PCs. Geologists arenow able to plot their own test plots, informational maps,and posters directly from their desks. Because they nolonger need to go through someone that knows the Sunsystem, it has encouraged more experimentation with datathan we had seen before.

WHERE WE ARE GOING (WE THINK!)

The hot topics of conversation in the GIS/Geologyworld of late seem to be interactive maps and 3-D map-ping. We are currently looking into putting three-dimen-sional views of some of the Aleutian volcanoes on theAlaska Volcano Observatory web page(http://www.avo.alaska.edu). We have also had some dis-cussion of 3-D geologic mapping as a part of our futureplans. It is difficult to try to envision exactly what thesenew technologies will mean to our business but one thingis certain. The trend is away from the traditional two-dimensional paper map product in favor of data packaged

in digital format. Whether this means digital data on com-pact disks, maps served over the Web, or a combination ofthese and other, yet to be conceived, delivery systems,remains to be seen.

The question now is – what’s the next step? We’vemade some headway in the past year. We are confidentthat many of the questions asked in this paper will be dealtwith in the near future. Many of the goals we set and dis-cussed after the workshop last year are still valid and someare becoming a reality, but what is next? What should werealistically look at as the next level in building our data-base and mapping program? We face some very real limi-tations in the future due to an oil-price driven state budgetcrisis. It is more important than it has been in over adecade that we are as efficient as possible and that wemake that efficiency apparent to the people that we serve.The dilemma becomes how to do this and still keepabreast of the advances in technology.

REFERENCESDavidson, Gail, 1998, Can We Get There From Here?

Experiences of the Alaska Division of Geological andGeophysical Surveys, in D.R. Soller, ed., Digital MappingTechniques ’98 – Workshop Proceedings: U.S. GeologicalSurvey Open-File Report 98-487, p. 13-14<http://pubs.gov/openfile/OF98-487/davidson.html>.

147

INTRODUCTION

The Geological Survey of Alabama (GSA), in cooper-ation with and with funding from the Alabama Departmentof Environmental Management (ADEM), is producing aseries of interactive CD-ROMs that provide an assessmentof the vulnerability of ground-water aquifers in Alabamato potential contamination from surface sources. For thepurposes of this project, the state has been divided into 13areas and a CD-ROM will be developed for each. TheCD-ROM for the first area to be studied, which includesthe coastal counties of Alabama (Mobile and Baldwin), hasbeen compiled as a prototype for the series. The CD-ROM for each area will contain an interpretive report,including text, tables, figures, and plates, developed inAdobe Acrobat format; a complete suite of the GIS datasets used for, or produced as part of, the project; and otherpertinent and useful data layers (United States GeologicalSurvey (USGS) Digital Raster Graphics (DRG) topograph-ic maps for the area, roads, streams, etc.). All GIS dataproduced for this project are being developed as ESRI,Inc. ArcView shapefiles, using ArcView 3.1.

The Digital Geologic Map of Alabama (1:250,000-scale) is being used as a primary source of data for theaquifer vulnerability project, and the recharge areas for themajor ground-water aquifers in each area are being derivedfrom this GIS data set. A second derivative layer delineat-ing the level of aquifer vulnerability to surface contamina-tion will be produced for each area.

Along with the interpretive report and the GIS data,each CD-ROM will contain ESRI, Inc.’s free GIS tool,ArcExplorer, and the Adobe Acrobat Reader. Thus, userswill be able to interact with the included data regardless ofthe availability of a commercial GIS software package.

A key element to the successful production of the CD-ROMs for this project is the conversion of map layouts,

especially plate-size maps (24” x 36” and larger), createdfrom the GIS data sets in ArcView, to high-quality, high-resolution Adobe Acrobat (.pdf) files for digital publica-tion. This paper will detail some problems that wereencountered during this conversion process and successfulsolutions that were employed to achieve the desiredresults.

PROJECT BACKGROUND

During the mid- and late-1980s, the USGS, in cooper-ation with ADEM, produced a series of reports that delin-eates the major aquifers in Alabama and characterizes theirvulnerability to surface contamination. To prepare thesereports, the state was divided into 13 geographic areas andeach of these was addressed in a separate report (e.g.,Mooty, 1988). These reports were widely disseminatedand used by ground-water resource professionals in thestate, as well as by the general public. Data for the reportswere compiled at the 1:250,000-scale using the best avail-able geologic maps to delineate aquifer recharge areas.For example, one source of data for the reports was the1926 Geologic Map of Alabama (1:500,000 scale), whichwas at that time was the most up-to-date and detailedstatewide geologic map available. Further, the USGSreports were compiled and published before the wide-spread usage of GIS technology and thus were developedas purely analog products with little consideration given topossible future digital conversion.

In 1988, the GSA published the first statewide geolog-ic map of Alabama based on new mapping and compila-tion since the 1926 map. The publication of GSA SpecialMap (SM) 220, Geologic Map of Alabama (Szabo et al.,1988) was the culmination of decades of work by numer-ous geologists, including not only those affiliated with

Alabama Aquifer Vulnerability CD-ROM Project:Digital Publication of GIS-Derived Map Products

By Berry H. Tew and Ruth T. Collier

Geological Survey of Alabama420 Hackberry Lane

Tuscaloosa, AL 35486Phone: (205) 349-2852Fax: (205) 349-2861

e-mail: [email protected]@gsa.state.al.us

GSA, but a number of others involved in mapping thegeology of Alabama for various purposes. SM 220, pub-lished at the scale of 1:250,000, comprises the largest scaleand most accurate border-to-border depiction of Alabama’sgeology as presently understood. In 1996 and 1997, theGIS Group at GSA successfully undertook a digital con-version of the 1988 geologic map, thus providing a robustGIS data set of Alabama’s surface geology (Tew et al.,1998).

By the mid-1990s, it was widely recognized that theoriginal aquifer vulnerability reports, although havingserved their purpose well, were badly in need of revisionand update in order to remain useful. In addition to theavailability of a new geologic map for the state, numerousnew data related to aquifers and ground water in Alabamahad been collected in the ensuing years due to drilling ofnew water wells and various monitoring programs andresearch efforts. For example, the early 1990s saw thedevelopment of numerous wellhead protection programsaround the state designed to meet the requirements ofamendments to the Safe Drinking Water Act originallyenacted by Congress in 1974. The 1986 amendmentsdirected the U.S. Environmental Protection Agency tooversee the states’ development of plans and programs toprotect areas providing ground water to public water sup-ply wells or springs. ADEM administers Alabama’sWellhead Protection Program (WHPP) which requires ageologic and hydrologic evaluation, delineation of well-head protection area boundaries, and a potential contami-nant source inventory for public drinking water supplywells and springs. Thus, development of the WHPP result-ed in an abundance of new ground-water-related data formany parts of the state.

In 1997, personnel in the Ground Water Branch atADEM contacted GSA relative to development of a pro-ject to update and revise the original thirteen aquifer vul-nerability reports. During the course of discussionsregarding the project, several desirable outcomes wereenumerated. First, it was decided that the 13 geographicareas of the original reports were familiar to the user baseand should be retained as the basic format for preparationof the new reports. Second, it was determined that datafor the reports should be compiled at the 1:100,000 scale,where possible, rather than at the 1:250,000 scale of theoriginal reports. Third, it was agreed that all data for theproject should be developed digitally in a GIS environ-ment, should result in useful, comprehensive GIS databas-es for ground water resources in Alabama, and should befully documented with Federal Geographic DataCommittee (FGDC)-compliant metadata. Fourth, it wasdeemed necessary that data in individual data layersshould be consistent and seamless from area to area so thateach layer would eventually constitute border-to-bordercoverage for Alabama at project completion. Fifth, it wasdecided that the resulting interpretive reports, GIS data,and metadata should be compiled and distributed in CD-

ROM format and that the CD-ROMs should provide inter-activity between the data and the user. Sixth and finally, adetermination was made that the CD-ROM should includea GIS data browser so that the user would not be requiredto have access to commercial GIS software to interact withthe included data.

To facilitate the desired project outcomes, a projectplan was developed by GSA and presented to ADEM. Theprimary elements of the plan included the following: eval-uate, revise, and recompile data from the original reportsas appropriate; compile new data; develop necessary GISdata layers, including incorporation of existing digitalgeospatial data where appropriate; develop interpretivereports on the basis of spatial analysis and hydrogeologicresearch; and, publish the reports and geospatial data on aninteractive CD-ROM. This project plan was approved andthe project was initiated in the fall of 1997.

DIGITAL PUBLICATION OF GIS-DERIVED MAP PRODUCTS

As indicated above, GIS data layers associated withthe aquifer vulnerability CD-ROM project are being devel-oped as ESRI, Inc. ArcView( shapefiles owing to the factthat ArcView( GIS software is in widespread use amongground-water professionals and many others in Alabama.Thus, developing data in shapefile format alleviates manyissues associated with data transfer and ease of use.Consequently, the majority of illustrations for the interpre-tive aquifer vulnerability reports, primarily consisting ofmaps, are being developed using the page layout tools inArcView. These maps range from rather simple, page size(8.5” x 11”) illustrations depicting one or two data themesto complex, plate size (24” x 36”) maps that illustrate mul-tiple data themes and the results of various spatial analysesand contain large areas of color and pattern fill. Althoughthe data used to generate the maps are being provided onthe CD-ROM, allowing the user to interact with the datafor their own applications and visualizations, providing ahigh quality, high resolution, static version of each mapgenerated by the research hydrogeologists at GSA as partof the Adobe Acrobat-formatted interpretive report forscreen viewing and printing is desirable. Thus, it is neces-sary to convert ArcView layouts into Adobe Acrobat docu-ments for integration in the reports. As we prototyped theproject, we discovered that this conversion process is notentirely straightforward.

The most direct method to create an Acrobat .pdf filefrom ArcView is to print from the ArcView layout to theAdobe Acrobat PDFwriter, a virtual Postscript printer thatis included with the Adobe Acrobat distribution. Usingthis method, print resolution is user-selectable up to 600dpi. Basically, ArcView creates a Postscript file from thelayout and sends it to the PDFwriter. The next-most directmethod is to export a Postscript file from ArcView and


then use Acrobat Distiller, a .pdf creation utility includedin the Acrobat distribution, to “distill” the Postscript fileinto an Acrobat file. Again, resolution is user-selectableup to 600 dpi. Both of these methods result in beautiful,high-resolution .pdf versions of the original ArcView lay-outs and either method is excellent for the creation ofAcrobat documents from small, relative simple layouts.However, owing to the manner in which ArcView writesPostscript, large and complex layouts such as the plate-sizemaps from our project take an inordinate amount of timeto draw (and redraw after zooming, resizing, etc.) on thescreen. An Acrobat .pdf file created from a typical plate inthe aquifer vulnerability project can take two minutes orlonger to fully redraw. We explored taking the ArcViewgenerated Postscript files through a “middleware” proce-dure, such as opening the file in Corel Draw or other soft-ware packages and printing to the PDFwriter, as well asusing the Save As command and “distilling” the resultingfile through Acrobat Distiller, but these procedures didnothing to eliminate the problems with screen draws andredraws. The lengthy draw times associated with Acrobatfiles created from ArcView-generated Postscript weredeemed unacceptable for our purposes and we set abouttrying to develop a workaround procedure that provided areasonable compromise between file resolution and draw-ing time.

The first procedure that we explored involved usingan export option from ArcView other than Postscript. Theidea was to export in a format (such as JPEG) that couldbe read by an intermediate software package (such asAdobe PhotoShop) that could then, in turn, be used to pro-duce the Acrobat documents using the PDFwriter. Thismethod produced less than desirable results due to theArcView’s limitations for export resolution in most fileformats. For most formats, such as JPEG and WindowsMetafile, the maximum resolution for export fromArcView is 144 dpi. At this resolution, plate-size maps areentirely too “jaggy” in appearance and, for the most part,barely legible. However, drawing times for the resulting.pdf files are extremely rapid.

On the basis of the experiences with the above proce-dures, it was determined that the most likely chance ofsuccess lay in identifying a software application with thecapability to directly open a high-resolution ArcViewPostscript file and save this file in an image format, suchas JPEG, at a relatively high resolution, which could thenbe opened and printed to Acrobat format using thePDFwriter. We were aware that Aladdin GhostScript, afree software package(<http://www.cs.wisc.edu/~ghost/aladdin/get550.html>),had the ability to work with, and convert between, a num-ber of file types and so we decided to experiment withGhostScript to determine if its capabilities suited ourneeds. We acquired a copy of GhostScript 5.50 forWindows NT and its graphical interface module, GSView2.70, and immediately determined that the package could

open an ArcView PostScript file. Further, GhostScript hasthe capability to “print” the open file to a 300 dpi JPEGfile. We worked through this procedure with one of thelarge, complex maps from the aquifer vulnerability projectand opened the resulting file in Adobe PhotoShop LE.The quality of the JPEG image was excellent. However,file size was rather large owing to the fact that theGhostScript JPEG file contained red/green/blue (RGB)color. To reduce file size, the RGB file was changed toIndexed Color within PhotoShop LE. We then printed thefile to the PDFwriter and opened the resulting Acrobat filewith Acrobat Reader to test legibility and resolution, aswell as drawing time. The file opened quickly, redrewquickly, and the overall quality was quite good. Thus, theexperiment using GhostScript as a “middleware” applica-tion was successful. Subsequently, we have used this pro-cedure to produce Adobe Acrobat files from ArcView lay-outs with excellent results.

SUMMARY

The joint GSA/ADEM aquifer vulnerability CD-ROMproject required the digital publication of map productsgenerated from layouts in the ArcView 3.1 GIS softwarepackage as Adobe Acrobat .pdf format files. Direct con-version of ArcView layouts to Acrobat files, especiallywhen dealing with large, complex map compositions,proved to result in Acrobat files that were unacceptabledue to inordinately long draw and redraw times. Severalprocedures were explored in attempts to address this prob-lem, but most resulted in quick file draw time, but unac-ceptable maps due to low resolution. One procedure, how-ever, employing Aladdin GhostScript 5.50 for WindowsNT and Adobe PhotoShop LE as “middleware” applica-tions, resulted in Acrobat format files that opened andredrew quickly and were of high quality in terms of reso-lution. Thus, a successful methodology for digital publica-tion of large, complex ArcView map layouts in AdobeAcrobat format was developed.

REFERENCES CITED

Mooty, W.S., 1988, Geohydrology and susceptibility of majoraquifers to surface contamination in Alabama; Area 13: U.S.Geological Survey Water-Resources Investigations Report88-4080, 29 p.

Szabo, M.W., Osborne, W.E., Copeland, C.W., and Neathery,T.L., 1988, Geologic Map of Alabama: Geological Surveyof Alabama, SM 220.

Tew, B.H., Taylor, D.T., and Osborne, W.E., 1998, Production ofthe 1:250,000-Scale Digital Geologic Map of Alabama, inSoller, D.R., ed., Digital Mapping Techniques ‘98—Workshop Proceedings: U.S. Geological Survey Open-FileReport 98-487, p. 5-7, <http://pubs.usgs.gov/openfile/of98-487/tew.html>.

ALABAMA AQUIFER VULNERABILITY CD-ROM PROJECT: DIGITAL GIS-DERIVED MAP PRODUCTS 149

151

SUMMARY

Over the past year, the National Park Service (NPS)has initiated a geologic resources inventory (GRI) to docu-ment and evaluate the geology of about 265 national park(N.P.), national monument (N.M.), national recreationalarea (N.R.A.), national historic site (N.H.S.), and otherunits of the National Park System (Gregson, 1998).Geologic resource workshops were held for park units inColorado, and new user-friendly GIS tools were developedfor a pilot digital geologic map project at Black Canyon ofthe Gunnison National Monument (BLCA) and CurecantiNational Recreation Area (CURE). The NPS-developedcross section and legend text display tools are an integralpart of a standard geology-GIS model that is in develop-ment. The evolving geology-GIS model is based on theWashington State Arc/Info GIS data model (Harris, 1998)that is being adapted for ArcView GIS and extended toinclude components of the AASG/USGS Draft DigitalGeologic Map Data Model (GMDM) (Johnson, Brodaric,and Raines, 1998). Cooperative projects for NPS units inColorado, Craters of the Moon National Monument, andCity of Rocks National Reserve are planned or in work,and scoping workshops are scheduled for park units inUtah during 1999.

INTRODUCTION

Bedrock and surficial geologic maps and informationprovide the foundation for studies of groundwater, geo-morphology, soils, and environmental hazards. Geologicmaps describe the underlying physical habitat of many nat-ural systems and are an integral component of the geo-physical inventories stipulated by the National ParkService (NPS) in its Natural Resources Inventory andMonitoring Guideline (NPS-75) and the 1997 NPSStrategic Plan. The NPS Geologic Resources Inventory(GRI) is a cooperative endeavor among the NPS GeologicResources Division (GRD), NPS Inventory andMonitoring (I&M) Program (Natural Resource InformationDivision - NRID), U.S. Geological Survey (USGS), andindividual state geological surveys to implement a system-atic, comprehensive inventory of the geologic resourcesfor NPS units. The NPS Geologic Resources Inventory forthe 265 selected park units consists of four main phases:

1) a bibliography (GeoBib) of geologic literature andmaps,

2) an evaluation of park geologic maps, resources, andissues,

Geologic Resources Inventory for the National Park System:Status, Applications, and Geology-GIS Data Model

By Joe Gregson, Anne Poole, Steve Fryer, Bruce Heise, Tim Connors, and Kay Dudek

National Park Service1201 Oak Ridge Drive, Suite 350

Fort Collins, CO 80525Telephone: (970) 225-3559


[email protected][email protected][email protected][email protected][email protected]


3) the acquisition and production of digital map productsand information, and

4) a report with basic geologic information, hazards andissues, and existing data and studies.

STATUS OF GEOLOGIC RESOURCESINVENTORIES

The NPS GRD and I&M Program sponsored aBaseline Geologic Data Workshop in Denver in the fall of1997 to get input from NPS, USGS, state survey person-nel, and cooperators about basic geologic data needs thatcould be provided by the I&M Program. At the Denvermeeting, Colorado, Utah, and North Carolina were chosenas pilot project states to maximize cooperation among theagencies. The group discussed and adopted the four maininventory phases which are reviewed briefly below.

The GeoBib project is completing the initial phase ofdata collection for existing geologic resources (maps andliterature) in each NPS unit and publishing the data on theInternet (URL: http://165.83.36.151/biblios/geobib.nsfLOGIN: geobib read PASSWORD: anybody). In addi-tion, index maps showing the location of associated geo-logic maps have been prepared for the parks in Coloradoand Utah. In general, after map coverage for each park isdetermined, map products can be evaluated, and if needed,additional mapping projects identified and initiated.

Pilot geologic issues/map scoping workshops (withattendees referred to as Park Teams) were organized in1998 to evaluate the resources in Colorado parks and willcontinue with projects in Utah during 1999. Park Teamsevaluate existing maps for existing and potential digitalproducts and identify any new geologic mapping needs.New geologic mapping projects may be initiated on acase-by-case basis after careful evaluation of park needs,costs, potential cooperators, and funding sources.

GRI cooperators are also assisting with geology-GISstandards to ensure uniform data quantity and quality fordigital geologic maps. In addition to standardized datadefinitions and structure, NPS resource managers alsoneed user-friendly GIS applications that allow the digitalgeologic map products to “look and feel” like the originalpublished maps. Ongoing pilot digitization projects areproviding additional experience and test beds for the geol-ogy-GIS model.

Park workshops suggest several applications for parkresource management from an enhanced understanding ofthe parks’ geology. Examples include the use of geologicdata to construct fire histories, to identify habitat for rareand endangered plant species, to identify areas with cultur-al and paleontological resource potential, and to locatepotential hazards for park roads, facilities, and visitors.Digital geologic maps will enhance the ability to develop

precise hazard and resource models in conjunction withother digital data.

After completion of map inventories, a geologic reportsummarizing USGS, state, academic, and NPS geologicalliterature and data will complete the project for each of the265 park units. The geologic report content, format, anddatabase are still being developed.

Geologic Mapping and Digitizing Projects

The NPS I&M Program has cost-shared new geologicfield mapping for Zion National Park with the State ofUtah. Additional field mapping projects have been pro-posed for 1999 to complete the geologic maps for Bent’sOld Fort N.H.S., Curecanti N.R.A., Mesa Verde N.P., andYucca House N.M. A pilot project to digitize 4 USGSgeologic maps for Craters of the Moon N.M. has beencompleted, and the digitizing of Black Canyon N.M. andCurecanti N.R.A. geologic maps is in progress. MesaVerde N.P. has completed a digitizing proposal, and pre-liminary plans are to initiate digitizing projects in 1999 forall Colorado parks with completed geologic maps.

The NPS Geologic Resources Inventory is beingactively developed with the cooperation of USGS and stategeological surveys. However, many opportunities for pro-ject collaboration exist that have not yet been identified,and effective communication among cooperators is a keyfactor for success of the inventory. Another challenge ofinventory planning is the development of digital map stan-dards that are adaptable to diverse geological conditionsbut still provide quality, uniform products and firm guid-ance for map developers. Indeed, the diversity of geologicresources found in the National Park System will provide acontinuing challenge for effective project management.The National Park Service has identified GIS and digitalcartographic products as fundamental resource manage-ment tools, and the I&M Program and GeologicalResources Division are developing an efficient inventoryprogram to expedite the acquisition of digital geologicinformation for NPS units throughout the country.

GIS ISSUES AND IMPLEMENTATION -MAKING GEOLOGY USER FRIENDLY

One of the unresolved issues facing developers of dig-ital geologic maps and geology-GIS models is how toinclude unit descriptions, explanatory text, references,notes, cross sections, and the variety of other printed infor-mation that occur on published maps. This issue is partic-ularly important to the National Park Service because thereare few geologists employed at parks, and resource man-agers rarely have the GIS and geologic expertise needed todevelop a useful product from digital layers of polygons,lines, points, and associated tabular data. The overarchingdevelopment goal of the NPS I&M Program is to produce

digital products that are immediately useful to anyonefamiliar with their analog counterparts. For geologicmaps, this means that the map unit legend must be sortedand shaded appropriately by geologic age and that all tex-tual, graphical, and other information from the publishedmaps must be available interactively to the user. In short,the digital product must “look and feel” like its publishedsource.

Since NPS resource managers use GIS as a tool in awide array of collateral duties, the I&M Program is devel-oping most digital products in ArcView GIS. ArcViewinterfaces effectively with other software running on theMS Windows operating system, and a new approach usingthe Windows help software, a MS Visual Basic graphicsviewer program, the ArcView legend editor, and theAvenue script language has been developed to automatethe display and query of published map information in theGIS.

GIS Map Unit Legend

In ArcView, a theme legend can be edited by selecting(i.e., double-clicking with the mouse) it in the legend.Once in the legend editor, the Legend Type: is set to“Unique Values,” and the Values Field: is set to theG_AGE_NO field. The G_AGE_NO field is a numericfield used to sort the map units by geologic time and isequivalent to the GUNIT.AGE.NO field in theGUNIT.MAIN data file of the Washington State datamodel (WSM) (Harris, 1998) and the class_seq field in theAASG/USGS Digital Geologic Map Data Model (GMDM)(Johnson, Brodaric, and Raines, 1998). In the NPSArcView model, this field has been replicated in theGUNIT.DBF table to facilitate automating the legend andrenamed to G_AGE_NO to accommodate the dBase IVfield name limitation of 10 characters. Once the LegendType: and Values Field: are set, the individual Symbols areedited to match the published map colors, the geologicmap symbol, and the unit name. After the legend is com-plete, it is saved to an ArcView legend file (.avl exten-sion). In general, the 8.3 file naming convention is used tofacilitate file sharing across all platforms.

Automating Map Unit Descriptions and OtherTextual Information

In most GIS applications, the spatial database struc-ture does not facilitate the use of voluminous textual data.For example, in ArcView, the database text fields onlyaccommodate 254 characters (320 for INFO tables) whichlimits the ability to include lengthy map descriptions withthe spatial data. Several options are available in ArcViewto overcome this limitation including concatenating data-base fields, independent text files, linking to other data-base system files, and linking to a MS Windows help file.

After testing several options, NPS developers have beenimplementing the Windows help system.

The Windows help system begins with data input orimport of all the textual data from the published geologicmap(s). In the Black Canyon/Curecanti pilot project, mapdescriptions, references, notes, and other text were aggre-gated from eight geologic maps. The table of contents list-ed all of the map unit symbols and names (equivalent tothe class_label and class_desc fields of the AASG/USGSGMDM Classification Object Table) sorted by geologicage. Subsequent pages listed the map unit descriptionswith one unit per page (many units had multiple descrip-tions from different maps) and were paginated by geologicage. At the end, references and notes from each map wereentered on separate pages. Help context IDs, topic names,keywords, page numbers, and linking codes were added tothe pages which were saved as a rich text format (.rtf) file.Then, the rich text file was compiled into a Windows helpfile.

Once compiled, the Windows help file can be openedand used with almost any MS Windows software. Thetable of contents has each map unit symbol and unit name“hot-linked” to the descriptions, and each description ishot-linked to the references and notes. Using the built-inWindows help tools, users can jump instantly to the tableof contents, page through the age-sorted unit descriptions,search for keywords, or index the file and perform full-textsearches of the entire file. The Black Canyon/Curecantipilot project help file consists of more than 50 printedpages of information for more than 130 map units.Advantages of the Windows help file are that most textformatting, such as font, size, color, etc., are preserved inthe final product, many graphics and tables are also sup-ported, and the help system can be developed somewhatindependently of the digital geologic map.

In ArcView GIS, three Avenue scripts were written tofunction with a toolbar button to automate the Windowshelp file and call unit descriptions interactively from thegeologic map. The button tool is only active when thegeology theme is turned on. The user selects the map unithelp tool from the ArcView toolbar and clicks on thedesired map unit to view the associated unit description.Using the map unit symbol (GUNIT_SYM) and the corre-sponding help context ID, the Avenue routine loads theWindows help file and pages to the map unit description.Thus, the map unit descriptions and other text are instantlyavailable to the user of the digital map.

Automating the Geologic Cross Sections

Geologic cross sections are integral components ofmany published geologic maps and provide important spa-tial visualization tools to assist users with understandingthe mapped geology. The I&M Program has developed asimple interactive system for displaying cross sectionsusing ArcView and a MS Visual Basic (VB) graphics

GEOLOGIC RESOURCES INVENTORY FOR THE NATIONAL PARK SYSTEM 153


viewer program. The cross sections are scanned digitalgraphics files that ArcView can load and display via sys-tem calls to the VB graphics viewer program. This allowsthe user to interactively select the cross section(s) to view.With projects such as the Black Canyon/Curecanti pilot,the ability to quickly view some 28 cross sections through-out the area is a powerful asset toward understanding thearea’s geology.

To prepare the cross sections for viewing, the graphicsare first scanned at 100 dots-per-inch (DPI) and saved as adigital JPEG (.jpg extension) graphics file. The JPEG for-mat was chosen to allow the graphics to be served andviewed over the Internet in the future. Once again, the 8.3file naming convention is used to facilitate sharing acrossall platforms, and file names are based on the map seriesdesignation and the designated cross section on the map(e.g., GQ1516AA.JPG is the A-A’ cross section on theGeologic Quadrangle Map GQ-1516).

Although ArcView and the Avenue language provideseveral ways to display graphics and images, ArcView’scapabilities are inadequate for efficient viewing of crosssections that could be up to 6” x 48” in size. Therefore, asimple VB graphics viewer program was developed to pro-vide this capability. The viewer displays the graphics at100% with the ability to scroll from one end of the sectionto the other.

To automate the graphics display, the cross sectionlines were digitized into an ArcView shape file (e.g.,blcagsec.shp and associated extensions). Two text fieldswere added to the shape file table (e.g., blcagsec.dbf):GSEC_ID, which contains the section line (e.g., A-A’), andGSEC_FILE, which contains the complete path and file-name of the cross section graphics file.

In ArcView GIS, three Avenue scripts were written tofunction with a toolbar button to automate the cross sec-tions and call graphics files interactively from the geologicmap. The button tool is only active when the cross section

theme is turned on. The user selects the cross sectionviewer tool from the ArcView toolbar and clicks on thedesired cross section line displayed on the map. Using thecross section line and the corresponding filename, theAvenue script loads the graphics viewer and displays theselected section. Thus, the cross sections are interactivelyavailable to the user of the digital map.

DRAFT NPS GEOLOGY-GIS DATAMODEL

As discussed above, a standard NPS geology-GIS datamodel is being developed based on new user-friendlyArcView GIS tools and adapted from the Washington StateArc/Info geology data model (WSM) (Harris, 1998). TheNPS model will be further developed over the next year toinclude additional components of the USGS Draft DigitalGeologic Map Data Model (GMDM) (Johnson, Brodaric,and Raines, 1998). Although the model is currently struc-tured for Arc/Info and many details still need to be workedout, one important adaptation is to standardize the databasefield names to 10 characters or less that will port seamless-ly among Arc/Info, ArcView, and other software compati-ble with the Dbase IV format. Arc/Info coverage and tablenames have been shortened to 7 characters to accommo-date renaming the file name stem for look-up tables (e.g.,CODEUNT.INF to CODEUNT1.DBF) when the stem hasalready been used for the spatial attributes (e.g., CODE-UNT.PAT to CODEUNT.DBF). Files and field names thathave been added or changed from the Washington Statemodel (Harris, 1998) are marked with an asterisk (*). Thenew NPS model’s data themes are listed below followedby a selected data dictionary for the five themes that havebeen partially adapted to date (Fig. 1). Although muchwork remains, final development of the NPS geology-GISdata model should be completed within the next year.

GEOLOGIC RESOURCES INVENTORY FOR THE NATIONAL PARK SYSTEM

Figure 1. Simplified relationships among the database tables discussed and outlined in the text. Boldtype denotes database file names for Arc/Info (top) and ArcView (below). The tabular relationships arecoded with “m” for many and “1” for one. Related field names are in italics.

155


*CODEGLG poly Map units or main geologic spatial data containing both polygon data describing the map units and linear data describing the interface between those units (WSM GUNIT).

*CODEGLN line Map units or main geological spatial data represented as lines due to map scale limitations (WSM GUNITLN).

*CODEGPT point Map units or main geological spatial data represented as points due to map scale limitations (WSM GUNITPT).

*CODEFLT line Faults and their descriptions (WSM GFAULT).*CODEFLD line Linear fold axes and their descriptions (WSM GFOLD).*CODESEC line Cross section lines and their attributes (no WSM equivalent).*CODEATD point Attitude observation points and their attributes (WSM GATTUD).*CODEDAT point Age-date sample location points (fossil or radiometric age estimates) (WSM

GDTSM PL).*CODEVNT point Volcanic vents, eruptive centers, and lithologic descriptions (WSM GVENT).*CODEDIK line Individual lithologic dikes and lithologic descriptions (WSM GDIKE).*CODEDKS polyAreas of lithologic dikes too numerous to map as individual segments (WSM

GDIKESWARM).

Coverages/Shape Files Data Dictionary*CODEGLG (network coverage containing both arc (.AAT) and polygon (.PAT) attribution)

DATA FILE NAME: CODEGLG.PAT or CODEGLG.DBFITEM NAME WID-TYP DESCRIPTIONAREA 4 - F PERIMETER 4 - F *GUNIT_ 4 - B (WSM GUNIT#; automatically converted in shape file .dbf)*GUNIT_ID 4 - B (WSM GUNIT-ID; automatically converted in shape file .dbf)*GUNIT_IDX 6 - I unique number for each polygon (WSM GUNIT.ID)*GUNIT_SYM 12 - C age-lithology unit symbol (WSM GUNIT.LABEL.CD)*USGS_SYM 12 - C geologic symbol from USGS geologic map(s)*G_AGE_NO 4 - F number to age-sort map units (from *CODEGLG.INF table)*GMAP_ID 4 - I code for *CODEMAP.INF look-up table (WSM GMAP.ID)

*GUNIT_SYM (map unit symbol in ASCII text; WSM GUNIT.LABEL.CD, GMDM class_label)Age-lithology unit polygon labels. Item is also the key used to relate the *CODEGLG coverage withthe *CODEGLG.INF or CODEGLG1.DBF (WSM GUNIT.MAIN) file that contains additional geologicattributes for the polygons.

*GMAP_ID (geologic map source information)Unique integer value that relates to series and citation information of the published map in the*CODEMAP.INF or CODEMAP.DBF file.

DATA FILE NAME: CODEGLG.AATITEM NAME WID-TYP DESCRIPTION*FNODE_ 4 - B (WSM FNODE#; automatically converted in shape file .dbf)*TNODE_ 4 - B (WSM TNODE#; automatically converted in shape file .dbf)*LPOLY_ 4 - B (WSM LPOLY#; automatically converted in shape file .dbf)*RPOLY_ 4 - B (WSM RPOLY#; automatically converted in shape file .dbf)LENGTH 4 - F

Coverages/Shape Files (modified from Harris,1998)

Present work is focusing on renaming files to consis-tently use the 8.3 file name convention with the NPS parkunit alpha code (CODE) and to limit field names to 10characters or less.


*GUNIT_ 4 - B (WSM GUNIT#; automatically converted in shape file .dbf)*GUNIT_ID 4 - B (WSM GUNIT-ID; automatically converted in shape file .dbf)*GCNTCT_IDX 7 - I unique number for each arc segment (WSM GCNTCT.ID)*GCNTCT_TYP 1 - I code for types of polygon boundaries (contacts) (WSM GCNTCT.TYPE.CD)FLTCNT 1 - C flags lithologic contacts that are also faults*GMAP_ID 4 - I code for *CODEMAP.INF look-up table (WSM GMAP.ID)*GCNTCT_TYP (polygon boundary/geologic contact type code)

1 known location2 concealed location3 scratch boundary4 gradational boundary5 shoreline6 approximate location7 quadrangle boundary8 ice boundary9 inferred location

FLTCNT (fault contact)Y Yes, the fault is also a lithologic contactN No, the fault is not a lithologic contact

*CODEFLT (arc or line coverage)

DATA FILE NAME: CODEFLT.AAT or CODEFLT.DBFITEM NAME WID-TYP DESCRIPTION*FNODE_ 4 - B (WSM FNODE#; automatically converted in shape file .dbf)*TNODE_ 4 - B (WSM TNODE#; automatically converted in shape file .dbf)*LPOLY_ 4 - B (WSM LPOLY#; automatically converted in shape file .dbf)*RPOLY_ 4 - B (WSM RPOLY#; automatically converted in shape file .dbf)LENGTH 4 - F *GFAULT_ 4 - B (WSM GFAULT#; automatically converted in shape file .dbf)*GFAULT_ID 4 - B (WSM GFAULT-ID; auto-converted in shape file .dbf)*GFAULT_IDX 5 - I unique ID number for each fault (WSM GFAULT.ID)*GFLT_SEG_N 4 - I unique number for each fault segment (GFLTSEG.NO)*GFLT_SEG_T 3 - I code value used to differentiate fault types and characteristics of the fault at

the segment level (GFLTSEG.TYPE.CD)FLTCNT 1 - C flags faults that are also lithologic contacts*GMAP_ID 4 - I code for *CODEMAP.INF look-up table (WSM GMAP.ID)

*GFAULT_IDX (geologic fault ID)A geologic fault is commonly segmented as a result of intersecting different polygons. This item identifies an indi-

vidual fault regardless of the number of segments. This item is also a key used to relate the *CODEFLT coverage with the*CODEFLT.INF file that contains fault names and data.

GFLT_SEG_T (geologic fault segment type code)1 fault, unknown offset2 fault, unknown offset, approximate location3 fault, unknown offset, concealed4 fault, unknown offset, queried5 fault, unknown offset, approximate location, queried6 fault, unknown offset, concealed, queried7 And so on for 83 entries.

FLTCNT (fault contact)Y Yes, the fault is also a lithologic contactN No, the fault is not a lithologic contact

157


*CODEFLD (arc or line coverage)

DATA FILE NAME: CODEFLD.AAT or CODEFLD.DBFITEM NAME WID-TYP DESCRIPTION*FNODE_ 4 - B (WSM FNODE#; automatically converted in shape file .dbf)*TNODE_ 4 - B (WSM TNODE#; automatically converted in shape file .dbf)*LPOLY_ 4 - B (WSM LPOLY#; automatically converted in shape file .dbf)*RPOLY_ 4 - B (WSM RPOLY#; automatically converted in shape file .dbf)LENGTH 4 - F *GFOLD_ 4 - B (WSM GFOLD#; automatically converted in shape file .dbf)*GFOLD_ID 4 - B (WSM GFOLD-ID; auto-converted in shape file .dbf)*GFOLD_IDX 6 - I unique ID number for each fold (WSM GFOLDT.ID)*GFLD_SEG_N 3 - I unique number for each fold segment (GFOLDSEG.NO)*GFLD_SEG_T 2 - I code value used to differentiate fold types and characteristics of the fold at the

segment level (GFOLDSEG.TYPE.CD)*GMAP_ID 4 - I code for *CODEMAP.INF look-up table (WSM GMAP.ID)

*GFOLD_IDX (geologic fold ID)A geologic fold is commonly segmented as a result of intersecting different polygons. This item identifies an individ-

ual fold regardless of the number of segments. This item is also a key used to relate the *CODEFLD coverage with the*CODEFLD.INF file that contains fold names and data.

GFLD_SEG_T (geologic fold segment type code)1 anticline2 anticline, approximate location3 anticline, concealed4 anticline, queried5 anticline, approximate location, queried6 anticline, concealed, queried7 And so on for 42 entries.

*CODEATD (point coverage or shape file)

DATA FILE NAME: CODEATD.PAT or CODEATD.DBFITEM NAME WID-TYP DESCRIPTIONAREA 4 - F PERIMETER 4 - F *GATTUD_ 4 - B (WSM GATTUD#; auto-converted in shape file .dbf)*GATTUD_ID 4 - B (WSM GATTUD-ID; auto-converted in shape file .dbf)*GATTUD_IDX 5 - I unique number for each point (WSM GATTUD.ID)*GATTUD_CD 2 - I code for type of attitude measurement (WSM GATTUD.CD)*GATTUD_ST 3 - I azimuth of strike or trend (0-360 degrees clockwise from the north with dip

direction clockwise from strike direction)*GATTUD_DP 2 - I dip or plunge degrees from horizontal (GATTUD.DIP.ANG)*GMAP_ID 4 - I code for *CODEMAP.INF look-up table (WSM GMAP.ID)

*GATTUD_CD (observation code for structural attitude point)1 strike and dip of beds2 strike and dip of overturned beds3 strike of vertical beds4 strike and dip of beds, dip unspecified5 approximate strike and dip of beds6 horizontal beds7 strike an dip of foliation8 And so on for 24+ more entries.


*CODEDAT (point coverage or shape file)

DATA FILE NAME: CODEDAT.PAT or CODEDAT.DBFITEM NAME WID-TYP DESCRIPTIONAREA 4 - F PERIMETER 4 - F *GDTSM_ 4 - B (WSM GDTSMPL#; auto-converted in shape file .dbf)*GDTSM_ID 4 - B (WSM GDTSMPL-ID; auto-converted in shape file .dbf)*GDTSM_CD 2 - I code for age-dating technique (WSM GDTSMPL.METH.CD)*GDTSM_NO 3 - I unique code for each age sample location (GDTSMPL.NO*GDTSM_AGE 80 - C relative or absolute age of rock sample (WSM AGE)*GDTSM_REM 254 - C notes about a specific age-date sample (WSM REMARKS)*GUNIT_SYM 12 - C age-lithology unit symbol (WSM GUNIT.LABEL.CD)*GMAP_ID 4 - I code for *CODEMAP.INF look-up table (WSM GMAP.ID)

*GDTSM_CD (geologic sample age-dating methodology code)1 radiometric2 paleontologic

*GDTSM_NO (geologic data sample number)This item contains the sample number from the original source map. The number is used to reference additional

explanatory text found on the map.

*GUNIT_SYM (map unit symbol in ASCII text)Age-lithology unit polygon labels. Item is also the key used to relate the *CODEGLG coverage with the

*CODEGLG.INF file that contains additional geologic attributes for the polygon.

*GDTSM_AGE (age of rock sample)The age of the rock sampled expressed as a relative geologic age, such as Jurassic, or an absolute age, such as

5,000,000 years before present.

*GDTSM_REM (remarks)Notes about the sample, such as the specific technique(s) used to determine the age of the sample.

Other Coverages/Shape Files

The other coverages/shape files listed above havenot yet been evaluated or adapted for the NPS geology-GIS data model, but the renaming of files and data fieldswill follow a similar pattern.

Accessory Data Files

*CODEGLG.INF (look-up data file, WSM GUNIT.MAIN)DATA FILE NAME: *CODEGLG.INF or *CODEGLG1.DBFITEM NAME WID-TYP DESCRIPTION*GUNIT_SYM 12 - C age-lithology unit symbol (WSM GUNIT.LABEL.CD)*GUNIT_NAME 100 - C formal name of map unit, if any*G_REL_AGE 5 - C relative age code (WSM GUNIT.REL.AGE.CD)*G_SSCR_TXT 6 - C subscript from the map symbol (WSM GUNIT.SSCRPT.TXT)*G_AGE_NO 5 - N number to age-sort map units (WSM GUNIT.AGE.NO)*G_AGE_TXT 50 - C geologic time period of map unit (WSM GUNIT.AGE.TXT)*G_MJ_LITH 3 - C 3 char. major lithology code (WSM GUNIT.MJ.LITH.CD)*G_LITH_NO 4 - B code used to sort on lithology (WSM GUNIT.LITH.NO)

159


*G_LITH_CD 10 - I code used to describe lithology (WSM GUNIT.LITH.CD)*G_LITH_TXT 100 - C brief text describing lithology (WSM GUNIT.LITH.TXT)*G_NOTE_TXT 254 - C descriptive notes about the map unit (GUNIT.NOTE.TXT)*GMAP_SRC 100 – C GMAP_IDs of map source(s), if any (similar to GMDM source_id)

*G_MJ_LITH (map unit major lithology code)EXT extrusive igneousINT intrusive igneousMET metamorphicSED sedimentaryVAS volcanic and sedimentaryUNC unconsolidated

Example record from *CODEGLG.INF or *CODEGLG1.DBF (modified from Harris, 1998)GUNIT_SYM = Qvba(pc)GUNIT_NAME = Basaltic Andesite of Puny CreekG_REL_AGE = QG_SSCR_TXT = vbaG_AGE_NO = 1.00G_AGE_TXT = HoloceneG_MJ_LITH = EXTG_LITH_NO = 39.10G_LITH_CD = vbaG_LITH_TXT = basaltic andesite flowsG_NOTE_TXT = volcanic lava flows with interbedded soil horizonsGMAP_SRC = I-757; GQ-1082

*CODEMAP.INF (WSM GMAP and GMAP.TXT INFO look-up data files combined/modified with Map and Sourcetables from AASG/USGS GMDM)

DATA FILE NAME: *CODEMAP.INF or *CODEMAP.DBFITEM NAME WID-TYP DESCRIPTION*GMAP_ID 4 - I code for relating this table (WSM GMAP.ID, GMDM map_id)*GMAP_YEAR 4 - I compilation or publication year (WSM GMAP.COMPILE.YR)*GMAP_AUTH 254 - C map author(s) (WSM GMAP.AUTHOR.YR; GMDM map_author)*GMAP_ORG 100 – C organization that created or compiled the map (GMDM org_id)*GMAP_TITLE 100 - C complete map title (WSM GMAP.REF.TXT; GMDM map_title)*GMAP_SERIES 20 - C map series or organizational identifier (e.g., USGS GQ-1516)*GMAP_SCALE 7 - I source map scale denominator (WSM GMAP.SCL)*GMAP_PROJ 100 – C name or description of map projection (GMDM map_projection)*GMAP_REF 254 - C complete map citation in USGS style*GMAP_DESC 254 – C brief description of the map (GMDM map_desc)*GMAP_XMAX 7 – I eastern limit of map in decimal degrees (GMDM map_xmax)*GMAP_XMIN 7 – I western limit of map in decimal degrees (GMDM map_xmin)*GMAP_YMAX 7 – I northern limit of map in decimal degrees (GMDM map_ymax)*GMAP_YMIN 7 – I southern limit of map in decimal degrees (GMDM map_ymin)*GMAP_SRC 100 – C GMAP_IDs of map source(s), if any (similar to GMDM source_id)

*CODEFLT.INF (look-up data file)

DATA FILE NAME: CODEFLT.INF or CODEFLT1.DBFITEM NAME WID-TYP DESCRIPTION*GFAULT_IDX 5 - I unique ID number from *CODEFLT (WSM GFAULT.ID)*GFAULT_NM 60 - C fault name, if any (WSM GFAULT.NM)*GMAP_ID 4 - I code for *CODEMAP.INF look-up table (WSM GMAP.ID)

REFERENCESGregson, Joe D., 1998, Geologic Resources Inventory—Geologic

Resources Division, Inventory and Monitoring Program, inDavid R. Soller, ed., Digital Mapping Techniques ‘98—Workshop Proceedings: U.S. Geological Survey Open-FileReport OFR 98-487, p. 109-111,<http://pubs.usgs.gov/openfile/of98-487/gregson.html>.

Harris, Carl F.T., 1998, Washington State’s 1:100,000-ScaleGeologic Map Database: An Arc/Info Data Model Example,

in David R. Soller, ed., Digital Mapping Techniques ‘98—Workshop Proceedings: U.S. Geological Survey Open-FileReport OFR 98-487, p. 27-35, <http://pubs.usgs.gov/openfile/of98-487/harris.html>.

Johnson, Bruce R., Boyan Brodaric, and Gary L. Raines, 1998,Draft Digital Geologic Map Data Model, Version 4.2:Association of American State Geologists/U.S. GeologicalSurvey Geologic Map Data Model Working Group Report,May 19, 1998, <http://ncgmp.usgs.gov/ngmdbproject/>.


*CODEFLD.INF (look-up data file)

DATA FILE NAME: CODEFLD.INF or CODEFLD1.DBFITEM NAME WID-TYP DESCRIPTION*GFOLD_IDX 5 - I unique ID number from *CODEFLD (WSM GFOLD.ID)*GFOLD_NM 60 - C fold name, if any (WSM GFOLD.NM)*GMAP_ID 4 - I code for *CODEMAP.INF look-up table (WSM GMAP.ID)

161

163

As new technologies emerge, methods of producinggeologic maps and other informational products are beingrefined and updated. At the Alaska Division of Geologicaland Geophysical Surveys, maps are produced on-demandfrom either an electrostatic or an HP design jet plotter.Growth of the geologic database and increased availabilityof outside data sources have led to increased efficiency inthe production of geologic maps, derivative maps, andother products. The use of Arc/Info in conjunction withPC-based drawing and layout software has proven to be aneffective method of producing a high quality product.

Two examples of projects are displayed. InformationCircular 38, Volcanoes of Alaska was produced by DGGSin conjunction with the Alaska Volcano Observatory. Thebase map consists of hypsography from State of Alaska,Land Records Information Section (LRIS) Arc/Info grids,coastline from LRIS ArcInfo coverages, and bathymetry

from USGS open file maps, combined using Arc/Info. Thetext was written by USGS and DGGS personnel and laidout using Adobe Pagemaker. The geologic map of theHorn Mountains area in southwestern Alaska is the resultof field work carried out during the 1990, 1991, and 1992field seasons. The geologic data was digitized in Arc/Infoand converted to an EPS file. The surrounding text filesare each individual Adobe Illustrator EPS files. All ofthese files were then linked in a single Illustrator file, inlayers, with one EPS file per layer. The file was then sentto an HP plotter with the PostScript (optimized for porta-bility – ADSC) option chosen.

Over the past year, we have found new ways of com-bining Arc/Info map data with PC drawing and layout pro-grams. These methods allow us to publish maps withsophisticated layouts more quickly, while still retaining ourunderlying Arc/Info database.

Examples of Map Production from theAlaska Division of Geological and Geophysical Surveys

By Andrew M. McCarthy and Frank Ganley

Alaska Division of Geological & Geophysical Surveys794 University Ave., Suite 200

Fairbanks, AK 99709Telephone: (907) 451-5005


165

A new geologic map of the Monterey 1:100,000 scalequadrangle and Monterey Bay is being compiled by theCalifornia Division of Mines and Geology and MossLanding Marine Laboratory. The map is a digital databasewhich, for the most part, was digitized at 1:24,000 scaleusing ALACARTE, a menu-driven module developed bythe U.S. Geological Survey running on Arc/Info GIS soft-ware. This project was supported in part by the U.S.Geological Survey through its National CooperativeGeologic Mapping program, STATEMAP component.

The map was prepared using the technique describedby Wagner at DMT ’97. Onshore, existing geologic maps,mostly at 1:24,000 scale were digitized and tiled togetherto make up the 1:100,000 scale map. An earlier analogversion of the map was used as a guide to correct bound-ary problems between quadrangles. A DEM with 30 meterresolution was used for a shaded relief basemap using thetechnique described by Ralph Haugerud and HarveyGreenberg at DMT ‘98.

Offshore, a wealth of new geologic information wasavailable. These new data were overlain on new high res-olution bathymetry acquired by the Monterey BayResearch Institute converted to a shaded bathymetric base.

REFERENCES

Haugerud, Ralph, and Greenberg, Harvey M., 1998, Recipesfor Digital Cartography: Cooking with DEMs, in DavidR. Soller, ed., Digital Mapping Techniques ‘98 —Workshop Proceedings: U.S. Geological Survey Open-File Report 98-487, p. 119-126.

Wagner, David L., 1997, Digital mapping techniquesemployed by the California Division of Mines andGeology, in David R. Soller, ed., Proceedings of aworkshop on digital mapping techniques: Methods forgeologic map data capture, management, and publica-tion: U.S. Geological Survey Open-File Report 97-269,p. 35-38.

Geologic Map of the Monterey 1:100,000 ScaleQuadrangle and Monterey Bay: A Digital Database

By David L. Wagner1, H. Gary Greene2, Jason D. Little1, Sarah E. Watkins1, andCynthia L. Pridmore1

1California Division of Mines and Geology801 K Street, MS-12-31Sacramento, CA 95814

Telephone: (916) 324-7380Fax: (916) 322-4765


2 Moss Landing Marine LaboratoryMoss Landing, CA

167

ABSTRACT

The Cartography Section of the GeoscienceInformation Division produces geological maps encom-passing a variety of geo-scientific data. The CartographicDatabase Standards (CDS) ensures that this data is man-aged in an efficient manner for map production andarchival purposes. The benefits of CDS allow geologistand cartographers to integrate and update data seamlessly,provide consistent and uniform digital data to the geo-science community, and ensure that map production isportable between cartographers. While CDS forms theframework of digital geological data, the Geological

Mapping System (GEMS) is the tool cartographers use tocreate the geological maps. Essentially, GEMS is com-prised of programs and a graphical user interface thatadapts the cartographic database standards and utilizes thestrengths of Arc/Info in a map production environment.Benefits of such a system are to increase productivity byusing universal methods and applications, and lessen thelearning curve of software as complex as Arc/Info.Further information on CDS and GEMS, including down-loading possibilities can be obtained from<http://www.NRCan.gc.ca/ess/carto> in the Proceduresand Specifications section.

Geological Map Production at the Geological Survey of Canada

By Vic Dohar and Dave Everett

Geological Survey of Canada601 Booth Street

Ottawa, Ontario, CanadaK1A 0E8

Telephone: (613) 943-2693Fax: (613) 952-7308

e-mail: [email protected]@NRCan.gc.ca

169

ABSTRACT

Geologic maps are an invaluable source of spatial dataabout surface and subsurface geology, and are increasinglyused by non-geologists. Notably, most GIS-capableagency, industry, and education units seek geologic mapsand digitize or import them into their systems. Geologicmaps are a challenge to understand, and many well-trainedGIS analysts and technicians lack an understanding of thescientific methods behind the interpretations on a geologicmap. This can lead to misuse of the data, especially whenthe data are in digital form.

Based on the USGS model, we are designing geologicmap products that meet the needs and expertise of the GIScommunity. Digital geologic maps are more than digitizedlines, symbols, and polygons. Without explanation, thosespatial data have little value. The digital product (digitalgeologic map) consists of spatial databases (composed oflines, polygons, and symbols), and a digital explanationdatabase. The digital explanation consists of a hierarchyof tables that are linked to each other and are linked to thecomponents of the spatial database. The power of a digitalexplanation is giving the user the ability to “query” orsearch the database, and generate a derivative version ofthe map. For example, one can search and select parame-ters such as lithology, stratigraphy, origin, landform,weathering, relative age, radiometric age, and sources ofdata (field mapping). Authors of digital geologic mapsmay include other parameters such as magnetic polarity,soil series, and hydrologic characteristics. This system oforganizing the scientific information that comprises a geologicmap provides a new way to present geologic observations andinterpretations that will aid in the educated use of these data.

IGS DATA MODEL DEVELOPMENT

The Idaho Geological Survey (IGS) is developing aderivative model based on the proposed USGS data model(version 4.2). The IGS model uses the ARC/INFO GISspatial database as its core. IGS has several goals toachieve in the development of a Geologic Map DataModel and the user tools associated with it:

* The data model and the organization of the resultingdigital geologic maps should be driven by the sci-ence of geology.

* In addition to creating the map and writing a legendand explanation, the geologic map author shouldhave direct involvement in composing the data-base fields.

* To achieve the above goals, we need to develop aset of user-friendly input tools.

* The model should be flexible, expandable, andeventually transferable to the final model stan-dards proposed by the Data Model SteeringCommittee or its technical working groups.

* Easy-to-use tools need to be developed for queriesand low-level analysis in a GIS package likeArcView.

Development of the IGS model began in September1998. Currently we have a prototype which is beingmodified. A prototype query engine tool has also beenwritten.

A Digital Geologic Map Data Model Designed for GIS Users

By Loudon R. Stanford, Kurt L. Othberg, Reed S. Lewis, and Roy M. Breckenridge

Idaho Geological SurveyMorrill Hall, Room 332

University of IdahoMoscow, ID 83844-3014

Telephone: (208) 885-7991Fax: (208) 885-5826


171

The digital geologic map of the Harrodsburg 30 x 60minute quadrangle was compiled from 32 U.S. GeologicalSurvey 7.5-minute geologic quadrangle maps (GQ’s). TheGQ’s are products of a cooperative mapping projectbetween the U.S. Geological Survey and the KentuckyGeological Survey from 1960 to 1978. This map is a com-pilation of existing maps, and no additional geologic fieldwork took place. When there were problems in strati-graphic correlation between quadrangles, the best currentdata available were used to resolve these differences.

The geology of the mapped area consists of flat-lyingsedimentary rocks of Silurian, Devonian, and Ordovicianage. These rocks extend across the north-trendingCincinnati Arch, crossing Ohio through Kentucky toTennessee. The Jessamine Dome is a regional domalstructure along the Cincinnati Arch in central Kentucky,and is reflected in the outcrop pattern of the UpperOrdovician units in the center of the map. Three majorfault systems are also present in the Harrodsburg 30 x 60-minute quadrangle: the Lexington, Kentucky River, andIrvine–Paint Creek Fault Systems.

The individual 7.5-minute quadrangle maps (scale1:24,000) were vectorized and attributed in Arc/Info fromraster images of the stable-base Mylar composites, using asemi-automated data-capture technique. Compiling 32individual 7.5-minute maps into a 30 x 60-minute map(scale 1;100,000) required resolving significant problems,such as (1) correlating geologic formations across quad-rangle boundaries, (2) displaying nonuniform structure-contour horizons, and (3) resolving discrepancies in

Quaternary alluvium boundaries and inferred contacts.Resolving the differences between quadrangles was neces-sary for topological analysis in a geographic informationsystem (GIS). In addition, lithologic members mapped onthe individual 7.5-minute quadrangle maps that weredeemed too small to be mapped at a scale of 1:100,000were mapped and consolidated with their adjoining mem-bers at the formation level.

A scaled schematic digital cross section was created inArc/Info, using ARCPLOT’s SURFACEXSECTION com-mand on the grid lattices created from the available digitalelevation models (DEM’s) and the structure horizon gridcreated from the compiled structure contours. After theresultant two line sections were created, the structure-con-tour horizon was intersected with the DEM surface hori-zon, and the subsurface horizons were copied and pastedthe appropriate distance below the surface during the edit-ing process.

The final product consists primarily of an ArcViewlayout of the compiled formation polygon coverage, com-piled fault and structure-contour arc and annotation cover-ages, economic point and arc coverages (quarries and min-eral veins), city and town annotations, and associated cov-erages of digital line graphs (DLG’s) of hydrology, roads,railroads, powerlines, pipelines, and county boundaries.Included on the final layout are a stratigraphic column,base maps, text, and the digital cross section. The finallayout was created in ArcView on a 36 x 65 inch page lay-out and plotted on a Hewlett Packard DesignJet 755CMprinter.

Digital Geologic Map of the Harrodsburg 30 x 60-MinuteQuadrangle, Central Kentucky

By Warren H. Anderson and Thomas N. Sparks

Kentucky Geological Survey228 Mining and Minerals Resources Building

University of KentuckyLexington, KY 40506-0107Telephone: (606) 257-5500


173

In the seven years that digital geologic maps havebeen constructed at the Montana Bureau of Mines andGeology (MBMG) using Arc/Info, a major problem hasbeen providing base information that makes the mapsmore useful. Because of cost considerations, very fewgeologic maps are being produced by offset printing. Inthe past few years, most maps have been released as open-file reports. We are moving toward producing these, andpossibly additional map series, as plot-on-demand digitalproducts.

The current fundamental mapping scale at MBMG is1:100,000. In the past few years, digital line-graph data ofcultural features and water and digital public land surveysystem data became available at several mapping scales.However, the most important base feature for geologicmaps is topography. To date, we have used three differenttechniques to add topographic data to digital geologic mapproducts, each of which has associated advantages andproblems: (1) mechanical topographic contours, (2) rastertopographic contours, and (3) digital hillshade relief. Themethods, problems, and successes of these differentapproaches are summarized below.

(1) Mechanical topographic contours: A blacklinemylar plot of the geologic arcs and labels is overlain on ablackline mylar of a U.S. Geological Survey (USGS) topo-graphic quadrangle to produce a sepia mylar from whichdiazo prints are made. This method is a holdover from ourprevious printing methods for production of open-filemaps (Figure 1A). Major problems with this approach are:(a) having to use a large vacuum frame and separate lightsource rather than our roller-drive diazo print machine; (b)needing screened mylars of all topographic quadrangles;(c) the need for linework to be relatively heavy for ade-quate reproduction quality, and (d) the resulting lack of a

colored product. We never fully carried out this proce-dure.

(2) Raster topographic contours: A raster representa-tion of a USGS topographic quadrangle is acquired from adigital raster graphic (DRG) file or by in-house scanning(usually at 400 dpi) of a mylar of the USGS topographicquadrangle. The image is then layered with geologic data.The topographic data from DRG files need some process-ing to strip out some colors, change the colorizing schemeto shades of gray, and turn the white areas transparent.The raster data are rectified in Arc/Info, on our Unix sys-tem. Image processing is done on personal computersusing Adobe Photoshop and Macromedia Freehand.Geologic data have been added with or without coloredpolygons. Our current approach is to overlay coloredpolygons with gray-scaled topographic data, and then putgeologic arcs and annotations in black over the other lay-ers, and to print the product on our 360 dpi inkjet plotter(Figure 1B). The main problems with the products are (a)the busyness of the maps where unit labels and other dataare superimposed on base data, and (b) the lack of contrastbetween some polygon colors and the gray topographicbase data. These color maps are more useful than mapswith blackline geology on blackline topography of method(1), but lack some of the crispness and readability of mapsproduced with vector basemap data.

(3) Digital hillshade relief: A gray hillshade imageproduced from a digital elevation model (DEM) is com-bined with the geologic coverage to produce a shaded-relief geologic map. The hillshade is constructed inArc/Info from available DEM’s (mostly 30 m data withartifacts in western MT and three arc-second data in east-ern MT). Filtering and vertical exaggeration of the eleva-tion data are typically used. Topographic and geologicdata are composited as grids, producing three grids in the

Different Needs, Different Bases: Struggling withTopographic Bases for Digital Geologic Maps

By Larry N. Smith and Susan M. Smith

Montana Bureau of Mines and GeologyMontana Tech of The University of Montana

Butte, MT 59701-8997Telephone: (406) 496-4379


[email protected]


Map (A)

Map (B)

Figure 1. Examples of geologic maps produced by two methods of adding topographic data to geology layers. Contacts,faults, and axial traces of folds are shown by solid or dashed lines; concealed contacts, faults, and axial traces of folds aredotted. The grids are sections in the public land survey system.

Map (A) Example of a sepia mylar used to make diazo prints of Open-File Report, using method (1) to supply topograph-ic contour and geologic data. Note that (a) the thick line weights required by the diazo print process make for poor defini-tion between geologic and base data, and (b) the general busyness of the map.

Map (B) Gray-scale plot of a color print, made from a 360 dpi inkjet plotter, showing results of method (2). Gray-coloredbase data, on the original print, are on top of the colored polygons, but beneath the black arcs and annotations. Resolutionof some of the base data is poor; some of the labels on the base are illegible. There is an overall decrease in the busynessof the linework in comparison to Map (A).

hue-saturation-value model, which allows for lightening ordarkening of the image when a graphics file is created(Renaud, 1994). Where adequate DEM data are available(currently only in western MT), the products make visuallystriking displays that show geologic data in a highlyunderstandable format. However, the lack of topographiccontours hinders geologic applications that require quanti-tative elevation data.

Ideally, at a minimum we hope to print and providedigital versions of our geologic maps with contour bases;we hope also to provide hillshaded topographic bases,depending on the needs of the customer. Acquisition ofraster topographic contour data that is legible at mapscales is a surmountable problem; acquiring similar vectordata is more of a problem. Method (2) will likely be used

to produce our normal products. The enthusiastic recep-tion we have received for geologic maps on shaded relieftopographic bases encourages us to continue to producethese products.

REFERENCES

Renaud, 1994, Representing Color Spatial Data OverMonochrome Images: Proceedings of the FourteenthAnnual ESRI User Conference-May 1994.

DIFFERENT NEEDS, DIFFERENT BASES: STRUGGLING WITH TOPOGRAPHIC BASES FOR DIGITAL MAPS 175

177

INTRODUCTION

The North Dakota Geological Survey (NDGS) is theprimary source of geological information for the state ofNorth Dakota. The NDGS was created in 1895 for the pur-pose of identifying and cataloging mineral resources with-in the state. Over the years this mission has been expand-ed and now includes three primary responsibilities: 1)investigate the geology of North Dakota; 2) act in an advi-sory capacity to other state agencies and administer regula-tory programs; and 3) provide public service and informa-tion to the people of North Dakota. To fulfill theseresponsibilities, NDGS geologists conduct diverse studiesof the state’s surface and subsurface geology, publish theresults in a variety of formats, and provide cooperative andoutreach services to other government agencies, publicgroups, and interested individuals.

No single media has proven as effective as the geolog-ic map in disseminating geologic information to geologicprofessionals and the general public. Geologic maps inone form or another have been included in nearly allNDGS publications, from the first Biennial Report in 1901to our current STATEMAP projects. Over this period,mapping techniques and technologies have changed dra-matically. NDGS geologists have increasingly been utiliz-ing digital (ie. computerized) methods for map productionin an effort to more efficiently and economically publishthe results of our research.

HISTORY OF NDGS GIS CENTER

The NDGS took a lead role in geographic informationsystem (GIS) development in North Dakota. A GIS refers

to the computer software and hardware needed to input,store, analyze, and output geographic data. In a largersense, the “system” also refers to the trained personnel,data, and organizational structure needed to effectivelyconvey and use geographic data (Vonderohe, et al, 1991).The NDGS began developing its GIS in 1989 (Figure 1)after NDGS geologists were introduced to the technologyat a national meeting. The NDGS created a strategy forassembling the necessary hardware and software compo-nents and entered into an agreement with the U.S.Environmental Protection Agency (EPA), StateDepartment of Health and Consolidated Laboratories, andState Department of Agriculture to develop a GIS housedat the NDGS offices in Bismarck. The EPA provided thefirst hardware as an in-kind grant, including two DataGeneral Unix workstations, two Laser printers, and onecolor printer. With the purchase of Arc/Info GIS software,the NDGS established its GIS Center in 1991. A SUN IPXworkstation, a large scanner, a large digitizer, an additionalcopy of Arc/Info, and ArcView software were purchased in1992, thereby increasing GIS capabilities. The NDGS alsotook responsibility in 1991 for creation and maintenanceof the North Dakota GIS Spatial Data Clearinghouse.Most of North Dakota’s digital spatial data is managed anddistributed by the NDGS on the world wide web at<http://www.state.nd.us/ndgs/gis.html>.

Many of the first NDGS GIS projects produced mapsfrom previously existing spatial data sets. Early GIS pro-jects were a cooperative effort between the NDGS andother state agencies and included production of oil welllocation maps, endangered species maps (with the cooper-ation of the ND Game and Fish Department), and legisla-tive districts map (with the cooperation of the Secretary ofState’s office). Geographic Information System Center

Evolution of Digital Geologic Mapping at theNorth Dakota Geological Survey

By Karen J.R. Mitchell, Ann M.K. Fritz, and Ryan Waldkirch

North Dakota Geological Survey600 E. Boulevard AvenueBismarck, North Dakota

Telephone: (701) 328-8000Fax: (701) 328-8010

email: [email protected]@[email protected]

staff also made it a priority to convert existing geologicmaps to a digital format. The first product to utilize thepower of GIS modeling capabilities was the Shaded Reliefof North Dakota map (Luther et al., 1995), created fromdigital elevation models from the Defense MappingAgency (Figure 2).

Use of digital techniques spread to our active mappingprograms, including STATEMAP projects. STATEMAP, acomponent of the National Cooperative Geologic MappingProgram, is a cooperative program providing USGS fund-ing for geologic mapping completed by state geological

surveys. The first NDGS STATEMAP project to use digi-tal methods was the geologic mapping of the Jamestown,ND, area completed in 1994 (Biek, 1994). Three 7.5minute quadrangles were mapped at a scale of 1:24,000.Unfortunately, the project was hampered by the lack ofdigital base data available at that scale. Base data, likeU.S. Geological Survey digital line graphs, were availableonly at a scale of 1:100,000 for most areas of NorthDakota at that time and only one of the three quadrangleshad an existing 1:24,000 digital base. The nextSTATEMAP project, geologic mapping of four 7.5 minutequadrangles in the Dickinson, ND, area, was completed in1995 (Biek and Murphy, 1995). The Dickinson projectwas the first NDGS mapping project to use digital tech-niques for final preparation of all map products.

MAP PRODUCTION PROCEDURES

Geologic mapping projects completed since the GISCenter was established have utilized an approach whichcombines both manual and digital techniques. The proce-dure has changed little since the NDGS began digital mapproduction back in 1991. Field data are compiled on aconventional paper base map and interpreted to produce adraft geologic map. The geologic features are then draftedin ink onto a mylar positive of the base map and the inkedgeologic information is digitized in Arc/Info using a digi-tizing tablet. Features are attributed and coded to createthe final digital geologic map. After review by a cartogra-


1895: NDGS created

1901: First Biennial Report published

1956: First formalGeologic Map of North Dakota published

1962: First CountyBulletin published

1985: County Bulletin Series completed

1989: NDGS introduced to GIS technology

1991: NDGS establishes GIS Center

1994: Jamestown projectcompleted, first digitally produced geologic map

1995: Dickinson report and Shaded Relief Map of North Dakota published

1997: Begin conversion to NT-based GIS

1999: Pilot project using heads-up digitizing in geologic mapping

1998: Addition of Sun Ultra1 workstation

Figure 1. Timeline showing evolution of the geologic mapping program at the NDGS from the agency’s creationin 1895. The mapping program has been undergoing rapid change since introduction of digital technology in1989.

Figure 2. Shaded relief map of North Dakota. This map,published as NDGS Miscellaneous Map 32 (Luther et al.,1995), was the first NDGS product to utilize the modelingcapabilities of GIS.

pher and the mapping geologist, the digital map is readyfor release.

Final maps are available to the public as either astand-alone, print-on-demand (POD) product or as part ofa NDGS publication. The maps produced recently fromSTATEMAP projects have been published as open-filereports and have been printed for customers on demand.The maps are also available as digital data files upon cus-tomer request.

FUTURE MAP PRODUCTIONPROCEDURES

The NDGS digital mapping program continues toevolve as new technologies and techniques become avail-able. Desktop GIS software packages with increasingfunctionality and “heads-up digitizing” are an example ofthe new technology. In heads-up digitizing, the mappinggeologist transfers field data from their field maps directlyinto a digital format using a desktop (or workstation) GISsoftware package. The NDGS recently receivedSTATEMAP funding to begin a new geologicmapping/compilation project in northeastern North Dakotain which we intend to use ArcView v. 3.1 to enter geologicfield data directly from our field maps using U.S.Geological Survey Digital Raster Graphics as a base layer.Our approach is based on that used by the MissouriDivision of Geology and Land Survey (Starbuck, 1998).Heads-up digitizing eliminates the intermediate draftingand digitizing steps and should eliminate error introducedduring those processes. We hope this technique will alsoeventually improve map-production efficiency by reducingthe amount of time spent editing and reviewing the mapsprior to publication.

DISCUSSION

Our shift to heads-up digitizing, desktop GIS, andPOD products is in response to increased demand for digi-tal map products and the increased number of maps wewish to publish within a limited publications budget. Wemust face numerous issues as the NDGS GIS Centerevolves, including stability of our POD products, metadataand “version control” as we move towards greater use ofdesktop GIS, standardizing of map units and colors, andincreasing quality within the limitations of our budget.

Stability

The greatest advantage of the POD map productionprocess is that POD is a relatively inexpensive way to pro-vide information to the public. In addition, storage spacein our publications and map sales area is saved because ofreduced inventory. Disadvantages to our current POD

process include unstable media, increased waiting time forcustomers, and decreased quality. Currently we use aHewlett Packard DesignJet 2000CP with standard ink jetpaper. Printing sometimes takes 15 minutes or more andcustomers have to wait for their map to spool, print, anddry. The ink fades significantly in a relatively short time,especially if displayed in direct sunlight. The ink also hasa tendency to rub off with handling and is destroyed bymoisture, thereby decreasing the quality of the printout.Stability and quality can be improved by using a coated orglossy paper designed for printing high quality products.However, using a more expensive paper results in some-what higher POD cost.

Metadata Standards and “Version Control”

The GIS Center staff recently began using a standardmetadata template when constructing a new digital dataset. Metadata are the documentation about the reliabilityand quality of the source(s) of information used to create adata set. The NDGS metadata template is based on theUSGS template and complies with Federal GeographicData Committee (FGDC) standards. Due to limited staffresources, earlier data sets were not documented with for-mal metadata. In some cases, little information is avail-able to reconstruct complete metadata documents. We col-lect as much information as possible in these cases, andrelease them as “User Beware” data sets.

Use of GIS mapping techniques allows a user toquickly and easily update or change an existing data set.Although this is appealing, this speed and ease of changingdata sets can also be troubling for the GIS manager, espe-cially with increased usage of desktop GIS. At the NDGS,the most current version of a data set is stored in the GISCenter workstation, and all requests for digital informa-tion, either internal or external, must be made through theGIS manager. It is the GIS Manager’s responsibility to fillthat request and make sure that the user has the most cur-rent version of the data set. The data is transferred viainternal network, File-Transfer-Protocol, or a copy of thefile is made to CD-R or floppy disk. Desktop GIS usershave the responsibility to inform the GIS manager of anychanges they make to the coverages. It is hoped that thismethod will avoid possible duplication or accidental dele-tion of data sets.

Standard Map Units and Colors

We are fortunate that the geology of the entire state ofNorth Dakota has been mapped at a reconnaissance scaleof 1:125,000. The original maps were produced as part ofthe County Bulletin series, a cooperative effort betweenthe NDGS, ND State Water Commission, and the U.S.Geological Survey, that took 14 geologists 23 years tocomplete. However, the reconnaissance maps use a differ-ent color scheme for different counties and often use dif-

EVOLUTION OF DIGITAL GEOLOGIC MAPPING AT THE NORTH DAKOTA GEOLOGICAL SURVEY 179


ferent geologic names for similar geologic units. Thesevariations in colors and units were necessary adaptationsto our evolving understanding of North Dakota geologyand reflect the growth of geologic knowledge, not anyinadequacy of the mapping system. However, they dopose a challenge to our digital conversion efforts.

The NDGS has never established standard geologicmap units and colors for geologic studies in North Dakota.Currently, three geologists on our staff are working onSTATEMAP-funded projects. Each geologist creates alegend that is appropriate for the area in the state they aremapping and for potential users of the map. This lack ofstandardization creates problems when merging adjacentquadrangles mapped by different workers. The GIS man-ager, in consultation with the geologists, must reconcilethe differences between the adjacent quadrangles. This isoften a time-consuming process that should be avoidedwhenever practical.

FUTURE GOALS FOR NDGS GIS CENTER

The NDGS GIS Center will continue to evolve andgrow as we expand our use of digital techniques andacquire new technology to improve our products. We haveidentified seven goals to help us keep pace with this rapid-ly growing industry. Some of these goals address theshortcomings of our existing digital mapping program.Others are intended to enhance our programs and offeradditional products and services to our customers.

1. Explore printing techniques and materials toproduce higher quality print-on-demand products.

If we are to continue to produce most new geologicmap products as POD publications, we need to find eco-nomical ways to improve the quality and stability of theseproducts. Their usefulness is presently limited by their lackof durability.

2. Develop and maintain metadata for all NDGSdigital data sets.

Metadata need to be compiled for all existing NDGSdata sets. We will continue to adhere to our standardmetadata template for newly created data sets and revisethis template as necessary to maintain FGDC compliance.

3. Standardize map units and colors.The GIS Manager continues to stress the importance

of standardized map unit names and colors. The geologistscontinue to resist standardization and deal with edge-matching problems on a case by case basis. The mappinggeologists do make every attempt to reconcile their newmap with adjacent, preexisting maps. However, each geol-ogist brings their own unique perspective to interpretingfield observations. They prefer to create objectivelydefined units that suit the particular mapping application,rather than pigeon-holing their observations into a prede-termined standard.

4. Digitize all out-of-print maps and place them onthe world wide web.

In the past, a number of customers have requestedout-of-print maps. Currently, the best we can do is to referthem to the NDGS Library, the State Library, or a universi-ty library. As the use of digital data becomes more com-mon, having out-of-print maps in digital format will bebeneficial to not only our own agency, but also to otheragencies, such as the State Health Department, StateHistorical Society, ND Game and Fish Department, andState Water Commission.

5. Explore combined capabilities of GIS and strati-graphic software packages (Petra™).

The NDGS recently purchased Petra™, an integrateddatabase and stratigraphic software program. The fullpotential of this new software package combined with GISsoftware is yet to be explored at the NDGS. There aregreat potential benefits to all aspects of our research,including petroleum exploration, oil and gas reservoir esti-mates, geologic mapping, lignite reserve estimates, andother economic geology applications.

6. Hire an IT/GIS Analyst. An additional person with training and extensive

experience in computer networking, database management,and GIS management is necessary for the continued suc-cessful operation of the NDGS GIS Center. Currently, gen-eral computer maintenance is performed by an NDGSgeologist who happens to be very interested and knowl-edgeable about computers. Computer maintenance costscould be substantially reduced by hiring an IT/GIS analystto take over these duties from a staff geologist. TheNDGS has requested funding for this position, but at thetime of this writing, it remains to be seen if funding willbe approved by the state legislature.

7. Decrease turnaround time for map products.With the addition of GIS staff and the use of ArcView,

it is hoped that turnaround time for map production will bereduced. This will enable us to quickly produce mapsdealing with time-sensitive issues, like flooding or land-use changes, and make our agency’s products more visibleto the public and legislators.

CONCLUSION

The use of digital mapping techniques at the NDGShas greatly enhanced our mapping capabilities. Digitalmethods allow us to produce print-on-demand map prod-ucts that can be updated as new data become available orcustomized to serve a specific user need. This flexibilityenables us to better serve our customers, whether they aretrained geologists, engineers, land-use planners, or thegeneral public. However, there is always room toimprove. We recognize the problems in our currentapproach, particularly with regard to product stability and

standardization. New technologies continually becomeavailable to simplify or streamline digital processes, newapplications of digital mapping are created, and new meth-ods of integrating diverse data are constantly being devel-oped. We hope that, by critically examining our currentsystem and participating in events like this Workshop onDigital Mapping Techniques, we can direct our growth tomake the best use of limited resources while maintainingour high quality standards.

REFERENCES

Biek, R., 1994, Geology of the Jamestown, Bloom, andSpiritwood Lake Quadrangles, Stutsman County, NorthDakota: NDGS Open-File Report 94-1, 62p., 6 pl., 1:24,000scale.

Biek, R., and Murphy, E., 1995, Geology of the Davis Buttes,Dickinson North, Dickinson South, and LehighQuadrangles, Stark and Dunn Counties, North Dakota:NDGS Open-File Report 95-1, 64 p., 9 pl., 1:24,000 scale.

Luther, M., Bassler, R., and Jirges, H., 1995, Shaded Relief ofNorth Dakota: NDGS Miscellaneous Map 32, 1 pl.,1:1,000,000 scale.

Starbuck , E., 1998, Digital geologic mapping at the MissouriDivision of Geology and Land Survey using ArcView 3 andUSGS Digital Raster Graphics; in Soller, D.R., ed., DigitalMapping Techniques ‘98 - Workshop Proceedings: U.S.Geological Survey Open-File Report 98-487, p. 103-104,URL <http://pubs.usgs.gov/openfile/of98-487/brief.html>.

Vonderohe, A.P., Gurda, R.F., Ventura, S.J., and Thum, P.G.,1991, Introduction to Local Land Information Systems forWisconsin’s Future: Wisconsin State Cartographer’s Office,Madison, Wisconsin, 59 p.

EVOLUTION OF DIGITAL GEOLOGIC MAPPING AT THE NORTH DAKOTA GEOLOGICAL SURVEY 181

183

The Pennsylvania Geological Survey initiated thesinkhole inventory project in 1985 to systematically mapkarst features on a county-wide basis. Each hard-copycounty report includes a brief explanatory text and copiesof 7.5 minute quadrangle maps (scale 1:24,000) that showthe locations of sinkholes, surface depressions, surfacemines, cave entrances, and carbonate bedrock geology.Fourteen counties have been surveyed in detail and six othershave had a less detailed, reconnaissance survey completed.

The PC-based GIS software ArcView v. 3.1 is beingused to analyze the data. Themes are created to include: adigital raster graphics (DRG) 7.5 minute topographic basemap, a digital geologic base, geo-referenced karst data,and a box-style grid with grid cell spacings set at 500 feet

to cover the 7.5 minute base map. Using sum and joinfunctions, each unit cell containing karst point data is iden-tified, the number of features contained within the unit cellis counted, and a color that corresponds to a particularrange of numbers is assigned to that unit cell. The endproduct is a color-coded map that shows the density ofkarst surface features over a given area. The distributionof these features becomes visually apparent by showinghigh-density (bad) areas and low-density (good) areas. Byusing the identify function, additional information about aparticular point can be obtained.

This customized data can be utilized for a variety ofprojects ranging from permit applications to land-use plan-ning to spatial analysis.

The Integration of Sinkhole Data and GeographicInformation Systems: An Application Using ArcView 3.1

By William E. Kochanov

Pennsylvania Geological SurveyP.O Box 8453

Harrisburg, PA 17105-8453Telephone: (717) 783-7253


185

In 1980, the Pennsylvania Geological Survey (PaGS)published a new edition of Map 1, Geologic Map ofPennsylvania. This full-color map, which was compiledand printed at 1:250,000 scale, shows the areal distributionand extent of the bedrock geologic units in Pennsylvania.Since its publication, Map 1 has remained the most cur-rent, available source of bedrock geology for the entirestate and has had widespread use for regional studies. Theincreasing use of GIS technology in the past five years hasled to numerous requests from federal and state govern-ment agencies and the private sector for a digital versionof the map. In response to these requests, the PaGS hasconverted the geologic features depicted on Map 1 to adigital format. Two data sets were prepared usingARC/INFO software, one for geologic units and faults,and the other for dikes. The data sets contain 194 geologicunits, more than 12,000 polygons, and more than 30,000arcs.

Production of the data sets involved preparation ofscanned images of the 1:250,000-scale geologic units andfaults from the 1980 map. Because of inherent problemswith the 1:250,000-scale base map, scans were also madeof the source materials that were used to compile the stategeologic map in order to more accurately locate the geo-logic contacts in relation to streams, roads, and other basefeatures. The images were georeferenced and projected to

transverse mercator, the projection of the 1980 map.Vector files were edited on screen on a computer worksta-tion with the images in the background for reference.

In compiling the data set for the geologic units, thePaGS extensively modified an earlier data set of geologicformational contacts prepared by the U.S. GeologicalSurvey, Water Resources Division. The PaGS edited orredigitized all arcs, relocated and corrected polygon labelpoints, digitized all fault lines, and added several items tothe polygon and arc attribute tables. Polygon attributesinclude map symbol, name of geologic unit, age, and threelithology designations. Arc attributes include type of geo-logic contact line, fault line, and boundary line. The dataset for dikes was prepared entirely by the PaGS and con-tains only line attributes: map symbol, name, age, litholo-gy, and the type of dike line. The accuracy of the attribut-es and positioning of the arcs and polygons were checkedby preparing and proofreading numerous check plots, veri-fying the data on screen, and preparing a script in ArcMacro Language to ensure correctness of polygon attributes.

It is anticipated that the data sets will be of use for avariety of regional or statewide GIS applications, includingwater-resource and environmental studies, land use plan-ning, conservation and ecosystem management, mineralexploration, and industrial development.

The Making of the 1:250,000-Scale DigitalGeologic Map of Pennsylvania

By Christine E. Miles and Thomas G. Whitfield

Pennsylvania Geological SurveyP. O. Box 8453

Harrisburg, PA 17105-8453Telephone: (717) 783-7269


[email protected]

187

HISTORY OF GEOLOGIC MAPPINGIN REGION

Since 1974, the USGS has conducted geologicalinvestigations relating to the tectonic and earthquake histo-ry of the Charleston, SC region. Earlier studies, such asRankin (1977) and Gohn (1983), were a combination ofgeophysical, tectonic, and seismic investigations.Near–surface, and surface geologic maps were publishedby Force (1978a, b) and McCartan and others (1980).Mapping of the region at a small scale was completed byMcCartan, Lemon, and Weems (1984), and subsequentmapping efforts by Weems and Lemon provided a moredetailed coverage of the Charleston area at 1:24,000 scale(Weems and Lemon, 1984a, b, 1987,1988, 1989, 1993,1996, Weems, Lemon, and Chirico, 1997). Weems hascontinued mapping surficial and near-surface formations inand adjacent to Charleston County and it is this latesteffort which has included the advent of digital geologicmapping in the region.

SOUTH CAROLINA COASTALPLAIN/CHARLESTON COUNTY PROJECT

The goal of the project is to develop a comprehensivelarge-scale, digital geologic dataset of surface and near-surface geology and a smaller scale subsurface geologiccoverage for the Charleston county region. Thus the map-ping project was divided into two sections. The first wasthe conversion and continued development of a large-scalesurface geologic database for the area. The second was asubsurface stratigraphic framework for the same region.

SURFICIAL GEOLOGY DATABASEDEVELOPMENT

The large-scale surficial and near surface mappingproject consisted of three types of map development.First, seven printed maps of Charleston County 7.5 minutequadrangles were converted into a geographic informationsystem (GIS) database. Second, all new mapping in theregion was compiled on mylar separates for the develop-ment of new digital databases. Third, several datasets pre-viously compiled but unpublished, consisting of handdrafted linework on greenline quadrangles, needed to beconverted to digital format. Although the same processwas employed for all types of data conversion, printed ver-sions and unpublished compilations, the GIS staff experi-enced different challenges during the conversion steps.

The GIS staff employed a method of scanning, vector-izing, editing, and attributing for digital database creation.Although the process was basically the same for previous-ly published maps, unpublished compiled greenlines, andnew mapping compilations, the differences in the originalmedium required differences in methodology. Previouslypublished maps, which had map separates available, wereeasily scanned and compiled, whereas greenlines ofunpublished data were more difficult to work with.

Scans that were made directly from hand drafted com-pilations of greenline quadrangles contained written textleader lines and other extraneous information that pro-duced complex linework, much of which required deletion(Figure 1a). For new mapping, geologists drafted onmylar that was overlayed on plotted digital basemap files(Figure 1b). These digital basemap files were updateddatasets of topographic features and are available in

The Development of a Digital Surface and Subsurface GeologicMap Database of Charleston County, South Carolina

By Peter G. Chirico

U.S. Geological SurveyNational Center, MS 926A12201 Sunrise Valley Drive

Reston, VA 20192Telephone: (703) 648-6950



Arc/Info format through the South Carolina Department ofNatural Resources, Colombia, South Carolina at<http://water.dnr.state.sc.us/gisdata/index.html>. In thecase of published maps, photographic reproductions oflinework were procured for scanning. These black-linepositives of scribed linework were ordered from EasternRegion Publications Group of the USGS where some ofthe scribed map separates still reside. Scans derived fromthis medium were clean and contained only geologic con-tact information (Figure 1c).

The scanning process utilized an Anatech Eagle 4080ET large-format scanner, capable of scanning documentsup to 40 inches in width. Separates, blackline, and mylaroverlays were scanned at different resolutions to providethe best possible file for the vectorization step. Blacklinepositives are scanned at higher resolutions, between 600and 800 dots per inch (dpi). Hand drafted greenlines andmylar overlays generally were scanned at 400 dpi. Thethreshold setting on the scanner regulates the amount oflight passing through the medium and therefore affects theclarity of the resulting scan. This threshold setting wasaltered for each type of media to produce the best possibleraster image for vectorization. Blackline positives werescanned using a threshold of approximately 160–180,whereas the hand drafted media needed a higher threshold(190–210) to eliminate some background noise and pro-duce cleaner lines. The raster scans were stored in .tiff or.rlc file format. These raster data formats are accepted bythe GTX-OSR (GTX Corporation - Phoenix, Arizona) vec-torizing software.

Vectorization of the scanned raster data was a straight-forward process and resulting output files of vector lineworkwere saved as Drawing Exchange Format (.dxf) files. Dxf

files are easily imported into Arc/Info where the editingprocess occurs. The editing of vector linework to buildtopologically correct datasets of geologic polygons andlines was the most important and time comsuming step.Vector linework was projected into the same coordinatesystem as the digital basemap files (UTM 17, NAD 1927).

Digital basemap files of hydrography and hypsogra-phy as well as roads and miscellaneous transportation weredownloaded and used as boss layers or backcoverages(drawn on screen in background during the digital editingphase) for the digital compilation. The digital basemapfiles of hydrography were of particular importance as bosslayers for compiling surficial geologic maps becausehydrographic and topographic features such as marshes,swamps, mudflats, and islands needed to be copied direct-ly into the new digital geologic coverage. This was doneto ensure that geologic and hydrologic features wouldcoincide exactly on future plotted copies and for any sub-sequent GIS analysis. Hypsography (topographic contourlines) was used in a similar manner. Instead of directlycopying lines however, geologic contacts were edited tofollow hypsographic lines where required by geologic con-ditions. For example, geologic contacts describing alluvi-um and artificial fill were edited to follow the appropriatelines of the topographic backcoverage.

Because the utility of having geologic data in a digitalformat is the ability to view, query, and analyze the data, adatabase was developed separately and included a varietyof qualitative information that might be useful to datausers. A simple database was constructed in MicrosoftAccess for each of the units depicted on the map. Thisdatabase was saved as a .dbf file. The .dbf file format wasconverted into an Arc/Info database file or .info file

Figure 1. Examples of three different geologic data mediums that were digitized. A. Greenline with handdrafted data. B. New hand drafted mylar overlay. C. Black-line positive map separate.

A B C

through the dbaseinfo command. The .info file was thenjoined to the spatial database with the joinitem command.An explanation of those database items is contained inAppendix 1. An example of the database for surficial geo-logic map units follows:

AREA = 9504.188PERIMETER = 376.169CHAR_GEOL# = 6CHAR_GEOL-ID = 730MAP_UNIT = QwcID = 9UNIT = Wando FormationFACIES = clayey sand and clayDEPO_ENV = backbarrier (estuarine/lagoonal)RELATIVE_AGE = late PleistoceneMIN_AGE = 90000.00MAX_AGE = 110000.00ABSOLUTE_AGE = 90-110 kaLITHOLOGY = siltMODIFIER_1 = clayeyMODIFIER_2 = sandyMODIFIER_3 = very fine-medium

MODIFIER_4 = bioturbatedFRESHCOL1 = pale-yellowish-orangeFRESHCOL2 = medium-light-grayWEATH_COL1 = grayish-yellowWEATH_COL2 = dark-yellowish-orangeCOMPACTION = moderate

The final digital databases are stored in an Arc/Infoformat but are commonly displayed and distributed in anArcView project. Figure 2 illustrates the combined set of7.5 minute quadrangles for the northeastern part ofCharleston County, South Carolina.

SUBSURFACE GEOLOGIC DATABASEDEVELOPMENT

The subsurface geologic dataset is a collection ofauger holes, water wells, and core holes drilled and loggedthroughout the South Carolina coastal plain. A preliminarydataset of only the Charleston County subsurface data lay-ers was completed to demonstrate the organization of the data.

DEVELOPMENT OF A DIGITAL SURFACE AND SUBSURFACE GEOLOGIC MAP DATABASE

Figure 2. Surficial geologic maps in an ArcView project.

189

This database was developed for six units from theMaastrichtian to the middle Eocene age.

Data for each corehole includes gamma and electriclogs, and lithologic and fossil data. Geologic formationsare recognized at different depths throughout the core byanalyzing the available data. Depths to each geologic for-mation are entered into an Access database for each coreanalyzed. Structure contour maps of any surface can thenbe prepared through contouring the data for a particularformation. The database is stored in a relational format toprevent duplication of information and to simplify theprocess of contouring surfaces of selected geologic units.The main data table stores identifier information for eachcore hole, and examples are shown in Table 1.

Ancillary data tables are linked to this identifier tablethrough a common field (USGS_NO) and contain numeri-cal values that show the depth to the top and bottom of theunit and each unit’s thickness (Table 2). Several tables arecreated for each hole, each for a different geologic forma-tion encountered at a different depth in that hole.

Contour maps were developed for the subsurface geo-logic units using Arc/Info ver. 7.1.1. Structure contourmaps of the top and bottom of each unit as well as anisopach map showing unit thickness were developed.These contour maps were then stored as ArcView shape-files. The collection of these subsurface layers was storedin ArcView GIS for the traditional two dimensional dis-play structure contour maps. Additional work was thendone to include several cross sections to illustrate theregions stratigraphy. Originally these cross sections werecreated in Micrographix Designer as .dxf files. The .DXFfiles were then opened in an ArcView View to be dis-played (Figure 3).

FUTURE RESEARCH

With the acquisition of better desktop tools for presen-tation of complex subsurface datasets, the project scientistshope to better communicate this information to the publicthrough visualization software. ArcView’s 3D Analyst

allows limited capabilities to show subsurface geologicunits and their location with respect to the surface.

Future research and development in the use and appli-cations of digital geologic maps will be an important stepin the future of digital geologic mapping. Further studieswith this dataset include designing maps to include perti-nent quantitative data suitable for GIS analysis. Examplesfor surfical databases include detailed grain size informa-tion correlated across geologic units, and blow count ordensity data for geologic units for some general measure


USGS_NO SCWRC_NO NAME COUNTY LATITUDE LONGITUDE TOTALDEPTHCHN_635 16DD-y3 Town of Sullivans Island Charleston 32.76444 79.83278 2540.00000DOR-37 23CC-I1 USGS-Clubhouse Dorchester 32.88806 80.35917 2599.00000

Crossroads #1

Table 1. Example of the main data table for subsurface geologic information.

USGS_NO DATUM_ELEV DEPTH_TOP ELEV_TOP DEPTH_BASE ELEV_BASE THICKNESSBRK-089 32.00000 328.00000 -296.00000 490.00000 -458.00000 162.00000BRK-272 18.00000 541.00000 -523.00000 678.00000 -660.00000 137.00000CHN-011 10.00000 749.00000 -739.00000 836.00000 -826.00000 87.00000CHN-012 10.00000 752.00000 -742.00000 845.00000 -835.00000 93.00000

Table 2. Example of ancillary data tables containing information for a particular geologic formation

Figure 3. Geophysical logs and cross sections arestored as .dxf files in ArcView database.

of strength. Geologic map databases must be prepared toaddress problems of cooperating agencies. In CharlestonCounty, issues ranging from coastal erosion and stormsurges, land use planning for growing coastal communi-

ties, groundwater resource issues, and further studies relat-ed to earthquake hazard and liquefaction potential allrequire geologic map data to produce viable derivativeproducts.


APPENDIX 1. ARC/INFO DATABASE DESCRIPTION

Data Format of Database

COLUMN ITEM NAME WIDTH OUTPUT TYPE N.DEC 1 AREA 4 12 F 3 5 PERIMETER 4 12 F 3 9 CHAR_GEOL# 4 5 B -

13 CHAR_GEOL-ID 4 5 B - 17 TYPE 5 5 C - 22 ID 8 11 F 0 30 UNIT 30 30 C - 60 FACIES 30 30 C - 90 DEPO_ENV 40 40 C -

130 RELATIVE_A 25 25 C - 155 MIN_AGE 8 20 F 5 163 MAX_AGE 8 20 F 5 171 ABSOLUTE_A 15 15 C - 186 LITHOLOGY 15 15 C - 201 MOD_1 20 20 C - 221 MOD_2 20 20 C - 241 MOD_3 20 20 C - 261 MOD_4 20 20 C - 281 FRESHCOL1 25 25 C - 306 FRESHCOL2 25 25 C - 331 WEA_COL1 25 25 C - 356 WEA_COL2 25 25 C - 381 COMPACTION 15 15 C -

Definitions of Fields

TYPE: These unique designations are the geologicmap symbols used for the various geologic units. Theupper-case letters refer to the general geologic ages of theunits (Q=Quaternary, T=Tertiary). The lower case lettersrefer to the name, specific age, and (or) facies of the geo-logic units (for example, Qhs=Quaternary+Holocene+bar-rier sands; Qwf=Quaternary+Wando Formation+fossilifer-ous shelf sand).

ID: Consecutively numbered units.UNIT: Names of geologic units. Names with upper-

case letters are formally defined (for example, WandoFormation). Units with names using lower-case letters areinformally defined (for example,Ten Mile Hill beds, artifi-cial fill).

FACIES: General statement of the depositional envi-ronment and (or) lithology of a geologic unit. FACIESrepresents the appearance and characteristics of a sedi-mentary unit, typically reflecting the conditions of itsorigin.

DEPOSITIONAL ENVIRONMENT: Describes theenvironment in which the sediments originally accumulat-ed (for example, estuary,barrier island, marine shelf).

RELATIVE AGE: Age of the geologic units as iden-tified on the standard geologic time scale. Defined interms relative to other geologic units rather than in termsof years.

MIN_AGE: Numeric field containing the approxi-mate minimum ages of geologic units in thousands ofyears. This field incorporates the minimum age rangefrom the Absolute Age field but in a numeric form.

MAX_AGE: Numeric field containing the approxi-mate maximum ages of geologic units in thousands ofyears. This field incorporates the maximum age rangefrom the Absolute Age field but in a numeric form.

ABSOLUTE AGE: Ranges or values for approximateages of geologic units as determined from fossil ages,amino-acid racemization ages, and (or) radiocarbon ages(ka=thousands of years, Ma=millions of years).

LITHOLOGY: The physical characteristics of rocksor sediments as seen in outcrops, hand samples, or subsur-

191


face samples. Described on the basis of characteristicssuch as color, mineralogic composition, and grain size.Principal sediment types found in the geologic units.Sands consist primarily of quartz grains between 0.0625mm and 2.0 mm. Silts consist primarily of quartz grainsbetween 0.0039 mm and 0.0625 mm. Limestone/marldeposits consist of true limestone or limy clay (marl).Muck consists of variable percentages of sand, silt, andclay mixed with a high percentage of fine-grained organicmaterial. Descriptions of sediments in the LITHOLOGYand MODIFIER sections are qualitative visual descrip-tions.

MODIFIER 1, 2, 3, 4: These descriptive terms modi-fy the principal lithologic terms. They are listed in a hier-archical order according to predominant characteristics.Modifiers may refer to size of sand grains (for example,fine-medium), to admixtures of materials of contrastinggrain size (for example, silty, sandy, clayey), to specialcomponents (for example, fossiliferous, phosphatic), tograin-size sorting (for example, well-sorted, clean), or tostratification type or other fabric characteristics (cross-bed-ded, bioturbated).

Standard Geologic Grain Size Chart:Gravel: larger than 2.0 millimetersSand:

Very coarse: 1.0 to 2.0 mmCoarse: 0.50 to 1.0 mmMedium: 0.25 mm to 0.50 mmFine: 0.125 mm to 0.25 mmVery fine: 0.0625 mm to 0.125 mm

Silt: 0.0039 mm to 0.0625 mmClay: smaller than 0.0039 mm

FRESH COLOR 1, 2: Dominant (1) and secondary(2) colors of sediments in fresh exposures and subsurfacesamples. Color identification was visually determined inthe field from the Geological Society of Americas standardcolor chart.

WEATHERED COLOR 1, 2: Dominant (1) and sec-ondary (2) colors of sediments in weathered exposures.Color identification was visually determined in the fieldfrom the Geological Society of Americas standard colorchart.

COMPACTION: Qualitative field assessments of sed-iment compaction stated as “loose”, “moderate”, or“dense”. Younger sediments (Pleistocene and Holocene)tend to be loose or moderately compacted. Older Eocene,Oligocene, and Miocene sediments tend to be dense rela-tive to the younger deposits but still may be penetratedwith a knife blade using moderate pressure. (Weems andothers 1997).

REFERENCES

Force, L.M., 1978a, Geological studies of the Charleston, SouthCarolina area - Elevation contours on the top of the Cooperformation: U.S. Geological Survey Miscellaneous FieldStudies Map MF-1021-A, scale 1:250,000.

__________ 1978b, Geological studies of the Charleston, SouthCarolina area - thickness of overburden map: U.S.Geological Survey Miscellaneous Field Studies Map MF-1021-B, scale 1:250,000

Gohn, G.S., ed., 1983, Studies related to the Charleston, SouthCarolina earthquake of 1886 - Tectonics and seismicity:U.S. Geological Survey Professional Paper 1313, 375 p.

McCartan, Lucy, Weems, R.E., and Lemon, E.M. Jr., 1980, TheWando Formation (Upper Pleistocene) in the Charleston,South Carolina area, in Sohl, N. F., and Wright, W. B.,Changes in stratigraphic nomenclature by the U.S.Geological Survey, 1979: U.S. Geological Survey Bulletin1502-A, p. A110-A116.

McCartan, Lucy, Weems, R.E., and Lemon, E.M. Jr., 1984,Geology of the area between Charleston and Orangeburg,South Carolina: U.S. Geological Survey MiscellaneousInvestigation Series, Map I-1472, scale 1:250,000.

Rankin, D.W. ed, 1977, Studies related to the Charleston, SouthCarolina, earthquake of 1886 - A preliminary report: U.S.Geological Survey Professional Paper 1028, 204 p.

Weems, R.E., and Lemon, E.M. Jr., 1984a, Geologic map of theMount Holly Quadrangle, Dorchester and CharlestonCounties, South Carolina. U.S. Geological Survey GeologicQuadrangle Map GQ-1579, scale 1:24,000.

Weems, R.E., and Lemon, E.M. Jr., 1984b, Geologic map of theStallsville Quadrangle, Dorchester and Charleston Counties,South Carolina. U.S. Geological Survey GeologicQuadrangle Map GQ-1581, scale 1:24,000.

Weems, R.E., and Lemon, E.M. Jr., 1987, Geologic map of theMonks Corner Quadrangle, Berkeley County, SouthCarolina. U.S. Geological Survey Geologic QuadrangleMap GQ-1641, scale 1:24,000.

Weems, R.E., and Lemon, E.M. Jr., 1988, Geologic map of theLadson Quadrangle, Berkeley, Charleston, and DorchesterCounties, South Carolina. U.S. Geological Survey GeologicQuadrangle Map GQ-1630, scale 1:24,000.

Weems, R.E., and Lemon, E.M. Jr., 1989, Geology of theBethera, Cordesville, Huger, and Kitteridge Quadrangles,Berkeley County, South Carolina. U.S. Geological SurveyMiscellaneous Investigation Map I-1854, scale 1:24,000.

Weems, R.E., and Lemon, E.M. Jr., 1993, Geology of theCainhoy, Charleston, Fort Moultrie, and North CharlestonQuadrangles, Charleston and Berkeley Counties, SouthCarolina. U.S. Geological Survey MiscellaneousInvestigation Map I-1935, scale 1:24,000.

Weems, R.E., and Lemon, E.M. Jr., 1996, Geology of ClubhouseCrossroads and Osborn Quadrangles, Charleston andDorchester Counties, South Carolina. U.S. GeologicalSurvey Miscellaneous Investigation Map I-2491, scale1:24,000.

Weems, R.E., Lemon, E.M. Jr., and Chirico, P.G., 1997, DigitalGeology and Topography of the Charleston Quadrangle,Charleston and Berkeley Counties, South Carolina. U.S.Geological Survey Open File Series, OF 97-531, scale1:24,000.


195

In many common geographic information systems(GIS), real-world objects are represented as points, lines,or polygons. The spatial representation of the object (i.e.,the point, line, or polygon), in turn is often associated withtabular data. In Arc/Info, points, lines, and polygons arereferred to as features and are stored in coverages (themat-ic data layers). Tabular data pertaining to these features isstored in feature attribute tables.

The mathematical procedure for representing geome-try and spatial relationships of these features is calledtopology. According to Understanding GIS: The Arc/InfoMethod (Environmental Systems Research Institute, Inc.,1997), the three major topological concepts are 1) lines(arcs) are connected at nodes, 2) arcs that connect to sur-round an area define a polygon, and 3) arcs have directionand left and right sides.

Certainly, topology based on points, lines, and poly-gons adequately represents many geologic features, espe-cially if those features represent single observations. Forexample, a sample location may be represented by a point,or an outcrop location by a point or polygon. However,many geologic objects are not easily represented by a sin-gle feature. This is further complicated by the mechanicsof the topological data structure, which often undesirablyfragment lines and polygons by placing nodes at line inter-sections. Consider two cases. In the first example, a sin-gle fault is offset by an intersecting fault, creating twoindividual line segments. In the second example, abedrock geologic map shows multiple locations where asedimentary unit has been eroded to reveal an underlyingvolcanic unit. Although the mapping geologist may haveinterpreted the fault segments as part of one system or theindividual volcanic unit polygons to be part of a continu-ous unit at depth, this interpretation is obscured by frag-mentation by the software. Although the attribute tables

usually reflect this information, the visual integrity of theunit is compromised. Further, database size is negativelyimpacted, since individual entities have their own recordsin the feature attribute table. Does this system adequatelyrepresent our knowledge of the known geology? Or arethere tools that can help us both maintain the geologist’sinterpretation of the data and minimize storage require-ments?

Arc/Info provides two other feature types often over-looked by the geologic community: regions (which arebased on polygons) and routes (which are based on lines).Essentially, regions and route systems aggregate overlap-ping, noncontiguous, and nested features, allowing the userto model them as singular entities. There is no limitationon the number of regions or routes that can exist within acoverage. When a region or route is initially created, asubclass name is assigned. Each region or route subclassthen has its own feature attribute table. Further, region orroute subclasses can be aggregated into larger region androute systems. It is important to note that the addition ofthese features types does not take away the geologist’sability to attribute individual entities. Individual polygonscan still be attributed in the polygon attribute table. Routesystems are even more flexible, allowing the user toattribute individual sections within a route.

If the volcanic unit discussed above was incorporatedinto a region subclass, the database user would betterunderstand how the field geologist viewed the unit: asmultiple occurrences of a single unit. In addition, thedatabase size would be minimized as the attributes pertain-ing to each individual polygon are consolidated into one.Similarly, it may be advantageous to construct route sys-tems with thrust, normal, and strike-slip subclasses. Thatway, individual faults of each type can be placed in their

Using Regions and Route Systems to ModelCompound Objects in Arc/Info

By Michele E. McRae


Telephone: (703) 648-6349Fax: (703) 648-6953



respective subclass without losing the ability to attributeindividual fault segments.

In conclusion, regions and route have a wealth ofapplications in geologic data modeling. Although I haveprovided two cases of how they may be applied, theirinherently hierarchical nature raises many implementationquestions. Certainly, it is not advantageous to implementthem in every case where lines or polygons share attribut-es. How extensively should they be used? When you dodecide to implement them, how do you determine the fun-damental ‘building block’ (i.e., the least divisible geologic

object) of your region or route system? What are the rami-fications of exporting data with region or route systemtopology to other GIS software? These questions need tobe addressed in order to thoroughly exploit the utility ofthese powerful tools.

REFERENCES

Environmental Systems Research Institute, Inc., 1997,Understanding GIS: The ARC/INFO Method:GeoInformation International, p. 2-10.

197

In the map area in east-central Illinois, extensivesands and gravels are buried within a thick sequence ofglacial deposits. These sands and gravels constitute ahighly productive regional aquifer, supplying water tonumerous cities and communities. Through a collabora-tive effort between the Illinois State Geological Survey(ISGS) and the U.S. Geological Survey (USGS), the surfi-cial, glacial deposits have been mapped using newly-developed 3-D mapping methods. The project’s goalswere to: 1) create a database of key stratigraphic informa-tion; 2) develop new methods for computer-aided mappingand presentation of data; 3) produce an integrated set ofgeologic maps and 3-D views for each glacial unit; and 4)use the geologic maps to help build a regional groundwa-ter management tool.

The new mapping methods were described at theDigital Mapping Techniques workshop in 1998 (Soller,1998, and <http://pubs.usgs.gov/openfile/of98-487/soller4.html>). These methods produced, in raster format,a set of maps of the complex subsurface geology that isinternally consistent and that can form the geologic frame-

work for a regional groundwater flow model. Three-dimensional views of each map were developed, to aidvisualization. [Text, selected maps, and animated views ofthe images are available at <http://ncgmp.usgs.gov/ecill/>.]The maps and images were assembled into a formal USGSmap product according to the following procedure.

In the study area, eight stratigraphic layers weremapped. For each layer, maps of unit elevation and unitthickness were created in Arc/Info Grid. Elevation datawere imported into EarthVision, to create the 3-D images.The maps and screensaves of the images were converted toPostscript format. These files along with the photographsand text were imported into Adobe Illustrator where thecomposition of the formal USGS map product was com-pleted. Plots generated on a HP3000 inkjet plotter at a res-olution of 600 dpi served as proofs during the editingprocess, and as interim products. The high resolution ofthese interim plots made them highly effective tools forrapidly communicating the project’s results to scientificaudiences and to potential public- and private-sector coop-erators in future projects.

Production of USGS Map I-2669 —A Folio of Maps and 3-D Images

By David R. Soller, Susan D. Price, and D. Paul Mathieux


Telephone: (703) 648-6907Fax: (703) 648-6937


199

In the 1980’s, the geology shown on USGS Map I-1970-A, -B, -C, and -D was compiled according to tradi-tional methods, using ink on mylar. The draft map showedthe thickness and character of Quaternary sediments in theUnited States east of the Rocky Mountains, at a scale of1:1,000,000. The map area at this scale is more than 30square feet. The geologic information was contained onfour separate mylar overlays. The map contained a highlycomplex portrayal of the geology, and it was determinedthat conventional methods of map production would yielda substandard product because of difficulties in registeringcolor separations with peelcoats. In 1987, after discussionbetween author and cartographer, it was decided that theemerging technology of computer-based mapping mightoffer the precision needed to produce high-quality printingnegatives. Also, in discussions with potential users of themap information, it became clear that if the map were inGIS format, its utility would be significantly enhanced.

Methods for producing this type of map did not, how-ever, exist in 1987. Therefore, we sought and obtainedfunding through the USGS Director’s “sweepstakes”,which awarded “seed money” to selected projects that pro-posed to develop innovative methods in GIS and digitalcartography. In 1988-89, we developed a method for scan-ning the manuscript maps, manipulating the digital mapdata, and outputting the data as color separates for mapprinting. This method required a Tektronix 4991 autovec-torizing scanner, Arc/Info, and a Scitex Response-80 edit-ing workstation and film writer. Details of the methodwere published as a user’s manual (Soller and others,1990). Although the hardware specifications described inthe publication are no longer useful because of technologi-cal advances, the general approach and processing steps(e.g., methods for map registration) still are applicable.

In the early 1990s, the manuscript maps were scannedand edited in Arc/Info by the authors and by others. Datafiles then were submitted to the National MappingDivision for processing on the Scitex and for the genera-tion of negatives to make proofs. These negatives wereregistered to negatives of the conventional greenline basemap. Text and figures were produced as graphic files andplaced manually; when assembled, the package was sub-mitted for printing. The first of these maps, I-1970-A, waspublished in 1993 (Soller, 1993). The last map, I-1970-B,was published in 1998. The edited digital data also wereused to build a geographic information system (GIS) data-base for analysis. The digital data were published asUSGS DDS-38 (Soller and Packard, 1998), which isonline at <http://pubs.usgs.gov/dds/dds38>.

REFERENCES

Soller, D.R., 1993, Map showing the thickness and character ofQuaternary sediments in the glaciated United States east ofthe Rocky Mountains: Northeastern States, the Great Lakes,and parts of southern Ontario and Atlantic offshore area(east of 80o31’ West Longitude): U.S. Geological SurveyMiscellaneous Investigations Series, Map I-1970 A, scale1:1,000,000.

Soller, D.R., and Packard, P.H., 1998, Digital representation of amap showing the thickness and character of Quaternary sed-iments in the glaciated United States east of the RockyMountains: U.S. Geological Survey Digital Data SeriesDDS-38, one CD-ROM. <http://pubs.usgs.gov/dds/dds38>.

Soller, D.R., Stettner, W.R., Lanfear, K.J., and Aitken, D.S.,1990, A user’s manual for a method of map scanning anddigital editing for thematic map production and data-baseconstruction: U.S. Geological Survey Circular 1054, 38 p.

Production of USGS Map I-1970 — A Regional Map ShowingThickness and Character of Quaternary Sediments

By David R. Soller and Will Stettner


Telephone: (703) 648-6907Fax: (703) 648-6937


201

The Virginia Division of Mineral Resources has anongoing commitment to vectorizing both new and previ-ously published geologic maps in Virginia. Our goal is toprovide not only paper maps but also GIS layers. We havedigitized over half of our published maps and several pre-viously unpublished maps. In getting to this point, wehave had to deal with many issues. This paper will discussone of these issues, specifically the use of offset printingto produce the required initial quantity of a map.

OVERVIEW OF HISTORY ANDMETHODOLOGY

The Virginia Division of Mineral Resources (VDMR)became interested several years ago in creating digital geo-logic GIS layers and in reducing publication costs by usingdigital production methods. It was important to avoidredundancy of effort by digitizing a map only once whilemeeting both goals. When none of the existing softwaremet our needs, the Virginia Division of Mineral Resourceshelped to develop customized software.

The software is MS-DOS-based and runs on MSWindows systems. Data entry involves “heads-up” tracingof lines, creation of polygons, attribution of lines andareas, and entry of point data (strike and dip, well loca-tions, etc.) over scanned images. Undergraduate geologymajors have done the bulk of this work. DMR staff doesthe subsequent editing. All digital maps undergo the samedetailed scrutiny as used in final editing of any traditional-ly produced map. The time required to digitize existinggeologic maps varies with complexity and scale of the

original map ranging from 15 days for a 1:24,000 quadran-gle to 60 days for a complete 1:100,000 map.

We create a raster image of the geologic map fromvector files that replicate the original published (paper)map in all respects. Geology, in color, appears over a sub-dued topographic/culture base. The raster image can bedistributed on a CD-ROM, via the internet, or plotted onour HP Design Jet 2500 CP. Publishing a geologic mapimage on a Web page or CD-ROM has the advantages ofquick distribution and low storage costs. However, manyclients still want a paper copy and either do not have theability to print in color or want the entire map printed.Plotters allow printing of high-resolution large paper mapsand allow an agency to avoid storage costs for lowdemand maps.

OFFSET PRINTING TO MEET INITIALQUANTITY REQUIREMENTS

In Virginia, the Division of Mineral Resources isrequired to supply 45 copies of a newly published map tovarious library collections. After the initial sales of anoth-er 40 to 50 copies, the demand then drops to one copy permonth or less. We have found it to be impractical to printthe initial 80 to 100 copies using an inkjet printer. Ouralternative is to establish conventional offset printing as ameans of meeting the initial demand, since it is still themost economic way to produce larger quantities of a map.

In a cooperative venture with the USGS, a standardoutput raster image of a geologic map was sent to theUSGS printing shop. Adobe PhotoShop was used to con-

Offset Printing of Raster Image Files

By Elizabeth V. M. Campbell1 and Robert B. Fraser2

1Virginia Division of Mineral ResourcesP.O. Box 3667

Charlottesville, VA 22903Telephone: (804) 951-6343


2U.S. Geological SurveyNational Center MS 903

Reston, VA 22092

vert the raster file of the finished map from RGB (red,green, and blue) to CMYK (cyan, magenta, yellow, andblack), for ink trapping enhancements and color separa-tion. The CMYK plate-making negatives were written ona large-format film writer. Traditional four color printingmethods were then used. Experience shows that theprocess from color separation through printing requiresfewer than eight man-hours, including time on the press,and that the quality of the resultant print is as good as thatproduced by traditional manual cartographic methods.

Costs

In the past, we paid $6,000 - $25,000 for 700 copiesof a 1:24,000 quadrangle map depending on the map com-plexity and paper. We printed 700 copies partly becausewe divided the production cost by the number of copiesproduced in order to derive a selling price. Printing a larg-er number did not increase the printing costs as much as itdecreased the final selling price for the map.

However, eighty percent of the production cost ofprinting a color geologic map was the cost of scribing andmaking peel coats. By using a raster image, we only haveto pay the actual cost of printing. Therefore, the numberof copies needed to have a reasonable per copy sellingprice is lower. The minimum number of copies printeddepends on the type of printing press, the cost of setting upa print run and the type of paper.

In addition to lower production costs, we save timebecause we do not have to make line weight and colorguides, nor create mylars for structure and formation sym-bols. Another advantage is that we can proof the mapprior to sending it to the press because we are able to seehow the text, symbols, line weights, and patterns will lookon the printed map.

Problems Creating a Raster Image for OffsetPrinting

PPI

Some issues associated with offset printing need to beaddressed. Two of the most notable are resolution of thedigital image (pixels per inch or PPI) and color shifts.When determining PPI for a raster image, the questionusually comes down to resolution versus file size. Thehigher the resolution, the more clear and the less blockythe detail. If the image is to be published digitally or evenprinted on an inkjet plotter, the only limitation is the sizeof the image file, which can rapidly become unmanageableas the PPI increases. However, for an image going to anoffset printing press, there is another consideration. Athigher PPI’s too much ink is put on the paper causing themap to be dark. The general rule of printing is the PPIshould be two times the lines per inch (LPI). LPI changes

depending on the paper and the press. LPI generallyranges between 72 and 150. We have gotten good resultswith image resolutions of 250 and 300 PPI. We want tocreate only one final image file for each map that can bepublished digitally or printed on inkjets or run through anoffset printing press. For this reason, we settled on animage resolution of 300 PPI, which preserves sufficientdetail, produces a file of manageable size and produces agood-quality print (though a little dark).

Color Shifts

We are in the process of resolving the issue of colorshifts. The problem is that computer monitors createimages by using red, green and blue (RGB) light. Inkjetplotters/presses generally use cyan, magenta, yellow andblack (CMYK) ink. Some colors that can be created usingRGB cannot be duplicated using CMYK. These colors aresaid to be “out-of-gamut.” Out-of-gamut colors lookmuddy when printed because the position in color spacehas been shifted slightly to bring it within the CMYKgamut or range of colors. It is difficult to predict theappearance of a printed map that was created in RGB.

One solution is to use “Spot” colors in the printingprocess. Spot colors are specific inks created to produce aspecific color not in the CMYK gamut. Since the colorproduced is not the result of a combination of four colorsbut rather one ink with a defined composition, the color isalways the same. The classic use of spot colors is in thereproduction of logo colors like “3M blue” where it isimportant to the company that the color is the same everytime it is used. On geologic maps, red is perhaps one ofthe more important colors that is difficult to reproduce inCMYK. Red spot ink is used sometimes as a fifth color inCMYK printing to obtain a true red. Using spot colors canbe very expensive since an individual mask plate is neces-sary for every color.

Another solution is to only use colors that are in the“common gamut”. The common gamut is that range ofcolors common to both RGB and CMYK; therefore, noshift is necessary. Some programs display a gamut warn-ing and offer substitutes. It can be time consuming to gothrough every “out-of-gamut” color on a map and find asubstitute particularly if the image does not use a colorpalette. Furthermore, RIPing programs that translate animage into plotter language can cause color shifts. This ismost noticeable in the yellow, which frequently developsgreenish tones. It necessary to adjust the program’sgamma curve. Not all programs allow the user to do this.

The issue of color shifts is further complicated by theunreliability of using an inkjet plotter to proof colors foroffset printing. Most inkjets use dye inks while offsetprinting presses use pigmented inks. Dye inks are water-soluble with the color in solution tinting the water. Theink is somewhat translucent and is susceptible to fadingupon exposure to UV light. Pigmented inks, on the other


hand, are oil-based and the color is produced by tiny parti-cles of pigment suspended in the ink. Pigmented inks aremore resistant to fading and are more opaque than dyeinks. Pigmented ink is available for the HP DesignJet2000 series printers. We have noticed that blues in particu-lar are different when printed in dye ink and pigmentedink.

While it is possible to buy pigmented ink for inkjetprinters and calibrate a monitor and a RIPing program toapproximate the CMYK press colors, it is still only anapproximation which must be constantly maintainedbecause ambient light, warmth of the cathode tubes and amyriad of other things affect our perception of color.Because of the difficulty of proofing offset printing colorsahead of time, many people who are only interested in theprinted result use one of several specific color systems, forexample TruMatch® and Pantone®. Each of these sys-tems has a swatch book printed on an offset printing pressunder specific conditions. These swatch books allow youto see how a color will look when printed. Once youselect a color from the swatch book, you either enter thecode for the color or the CYMK values for the color. Themajor image processing programs like Adobe Photoshop,CorelDraw, and Canvas support these two systems.

The Division of Mineral Resources has not reached adefinitive solution to the color problem. We will probablyuse a combination of the solutions mentioned above todevelop a chart of defined colors to be combined with pat-terns for various formations in Virginia. Our goal is toproduce one final raster image for each map where theimage will look the same regardless of how it is displayedor printed. In order to discern different formations within

the same age, colors need to maintain their distinctivenesswhen they are printed. We are working with the USGSprinting office on this problem. Together, we have createdtwo TIF files: a RGB color chart and a CMYK color chart.The USGS is planning to print the CMYK color chart ontheir offset printing press. It is possible to send the colorchart TIF files through various combinations of RIPingprograms to various printers. Using these charts and per-haps the TruMatch swatch book and color system, we willcreate a standard set of colors for each unit following theinternational stratigraphic color scheme.

CONCLUSION

The Virginia Division of Mineral Resources has suc-cessfully created an acceptable offset-printed map from a300 PPI raster image file. Sending digital image files toan offset printing press enables us to economically printthe initial number of copies required to release a newlypublished map with significant savings in both time andmoney. We are planning to use this procedure for offsetprinting of digital files to publish all new maps.

ACKNOWLEDGEMENT

We like to acknowledge Lawrence G. Matheson ofthe Map Application Center, National Mapping Division,U.S. Geological Survey for his efforts in processing theimage files through the pre-press procedures and for hisguidance in color management/separation.

OFFSET PRINTING OF RASTER IMAGE FILES 203

205

The topography of the land surface is a direct result ofthe type of geologic materials that underlie the surface andthe geologic events and processes that acted on them.Therefore, the shape of the land surface can provideimportant clues to many aspects of the Earth‘s history.

In contrast to the dominant landforms in the moun-tainous regions of North America, where topographic reliefcan exceed 5,000 ft over relatively short distances, thecentral lowlands of the United States are characterized byrelatively low relief topography. Elevations in Wisconsinvary by only 1,500 ft across the entire state, and soWisconsin landforms tend to be smaller in scale and moredifficult to view from a regional or statewide perspective.Glaciated and unglaciated regions dominate Wisconsin’stopography, and there are topographic features that areunique to each region.

Throughout the past two million years, continentalglaciers repeatedly flowed across much of what is now thenorthern United States. All but the southwestern quarter ofWisconsin has been covered by ice at different times; mostof the glaciated part of the state was glaciated during theWisconsin Glaciation (Clayton and others, 1991). As theice sheets advanced and receded, features such as drum-lins, eskers, and moraines were formed. Glacial activitywas recent enough in many parts of the state to be thedominant influence on the modern landscape.

The effects of stream incision dominate the topogra-phy of southwestern Wisconsin, which was never glaciat-ed. Years of drainage have worn down the surface into ahighly dendritic network of water channels (Clayton andAttig, 1997).

When evaluating geologic history, a map that effec-tively displays the way in which topography differs fromarea to area is usually essential. Topography can be por-trayed a number of ways: contour lines, oblique or stereo

aerial photography, and shaded relief. Shaded relief mapsprovide a means of viewing topography that requires littleor no interpretation. As an analytical tool, these maps canprovide the geologist with a unique regional perspective oftopography. The ease with which a shaded relief map canbe interpreted also makes it an effective teaching tool. Theadvent of digital geographic information systems, alongwith readily available digital elevation data, has made thecreation of shaded relief maps fairly straightforward, notthe highly interpretive and onerous task it once was. Thesubtle nature of Wisconsin’s topographic variations, espe-cially in the glaciated regions, necessitates the use of ahighly detailed elevation model to derive an effectiveshaded relief map.

Until recently, statewide digital elevation data wereavailable only in 500-meter and 75-meter raster format;these data were too coarse to show the fine topographicdetail that dominates the landscape in the glaciated parts ofthe state. Complete 30-meter USGS DEM coverage of thestate recently became available. The WisconsinDepartment of Natural Resources (DNR) compiled theapproximately 1,200 30-meter DEMs that make up thestate into a seamless grid in Arc/Info.

To derive the hillshading on this map, I used the tech-niques described by Haugerud and Greenberg (1998).Their methods involve the use of the GRIDCOMPOSITEcommand with the HSV option in ARCPLOT. This com-mand makes use of the HSV color model, where hueranges from 0 to 360, saturation from 0 to 100, and valuefrom 0 to 100, as shown in Figure 1. Essentially, GRID-COMPOSITE combines three grids representing HUE,SATURATION, and VALUE into a composite image. Thecombination of the analytical power of GRID with themulti-band imaging flexibility of the HSV model results inan effective visual technique.

Quaternary Landforms in Wisconsin — How Hillshading CanBe Used To Accentuate Topographic Features

By Chip Hankley

Wisconsin Geological and Natural History Survey3817 Mineral Point RoadMadison, WI 53705-5100Telephone: (608) 262-2320



For this map I wanted the color to be dependent uponelevation. I decided to use a range from green at the low-est elevations to orange at the highest elevations.

I created a HUE grid where the HUE values stretchedlinearly from green at the lowest elevations to orange atthe highest elevation. I had to do the same with the SAT-URATION grid, only using the SATURATION values thatrepresented my colors. I used the following formulas inGRID:

<HUE_GRID> = [HLOW] - [HLOW - HHIGH]

* ((<DEM> - DEMMIN) / (DEMMAX - DEMIN))

and

<SATURATION_GRID> = [SLOW] - [SLOW -

SHIGH] * ((<DEM> - DEMMIN) / (DEMMAX -

DEMMIN))

where: [HLOW] is the Hue component of the HSV

color representing the lowest elevations;[HHIGH] is the Hue component of the HSV

color representing the highest elevations;[SLOW] is the Saturation component of the

HSV color representing the lowest eleva-tions;

[SHIGH] is the Saturation component of the

HSV color representing the highest ele-vations;

DEMMIN is the minimum elevation; and

DEMMAX is the maximum elevation.

In the HSV color model, color is mainly dependentupon the H and the S values (just as in CMYK, color ismainly a function of C, M, and Y). In this technique, theVALUE grid controls the hillshade effect. Consider thatfor any pixel of color specified by HUE and SATURA-TION, the VALUE component sets the amount of black.For instance, a HUE of 120 and a SATURATION of 50will specify a medium light green. If the VALUE is 0, theresultant HSV color will be black. Similarly, if theVALUE is 100, the resultant color will be the mediumlight green with no black as a part of the color. This is theeffect of shading – shaded areas are blacker than illuminat-ed areas.

Haugerud and Greenberg (1998) note that the contrastof most images produced by the HILLSHADE commandin Arc/Info is not suitable for use in this process. Theysuggest modifying the standard hillshade via the followingtechnique:

&describe <hill_shade_grid>xxg1 = (<hill_shade_grid> - [Value GRD$MEAN]) *

15 / [value GRD$STDV] + 95)

Because GRIDS in Arc/Info are always rectangular, Ihad to manipulate the NODATA cells so that they wouldnot print black. I found that the easiest way to do this wasto convert the NODATA areas of the HUE, SATURA-TION, and VALUE grids into a white color. In HSV,white is any H value, S = 0, and V = 100. Using condi-tional statements in GRID, I converted the values of theHUE, SATURATION, and VALUE grids to their whiteequivalent wherever the original DEM was NODATA.

The final step in this process was to convert the threegrids into a geo-referenced TIF image. The geo-referencedTIF draws faster than the three grid composite and can bedisplayed in ArcView. To create the image from my HUE,SATURATION, and VALUE grids, I first created RED,BLUE, and GREEN grids using the HSV2RED,HSV2BLUE, and HSV2GREEN commands. I then com-bined the three grids into a stack using the MAKESTACKcommand using the LIST option. Finally, I used theGRIDIMAGE command to convert the STACK into a TIFimage.

I have attached the AML I used for this process in theappendix.

REFERENCES

Arc/Info 7.2.1- ARCDOC, 1997 [Computer Program], availablefrom: Environmental Systems Research Institute, Redlands,CA.

Clayton, Lee, and Attig, J.W., 1997, Pleistocene Geology of DaneCounty, Wisconsin: Wisconsin Geological and NaturalHistory Survey Bulletin 95, 64 p.

Figure 1. HSV color model (reprinted with per-mission from “Specifying Color,” ESRI ArcDoc7.2.1).

Clayton, Lee, Attig J.W., Mickelson, D.M., and Johnson, M.D.,1991, Glaciation of Wisconsin: Wisconsin Geological andNatural History Survey Educational Series 36, (revised1992), 4 p.

Haugerud, Ralph, and Greenberg, H.M., 1998, Recipes forDigital Cartography: Cooking with DEMs, in Soller, D.R.,ed., Digital Mapping Techniques ’98—WorkshopProceedings: U.S. Geological Survey Open File Report 98-487, p. 119-126, <http://pubs.usgs.gov/openfile/of98-487/haug2.html>.

QUATERNARY LANDFORMS IN WISCONSIN—HOW HILLSHADING CAN BE USED

APPENDIX

/********************************/* HSV.AML/********************************/*/* COLORED HILLSHADE PROGRAM/*/* Use this program to generate a colored shaded relief/* image. Color will be a function of elevation./* The user is required to supply the HSV color equivalents/* of a LOW elevation color and a HIGH elevation color/*/********************************/*/* Primary Developers: Chip Hankley - GIS Specialist/* Wisconsin Geological and /* Natural History Survey/*/* Derived from work by Ralph A. Haugerud(1) and/* Harvey Greenberg(2)/*/* (1) U.S. Geological Survey at University of Washington, Box/* 351310, Seattle, WA 98195, [email protected]/* (2) Dept. of Geological Sciences, University of Washington,/* Box 351310, Seattle, WA 98195, [email protected]/*/********************************/*/* Date of Initial Coding: 8/17/98/* Date of Last Edit: 3/25/99 /* Combined several different AMLs into one coherent module/* using routines. This should run on all platforms./*/*/********************************/*/* Available from: ARC/* Arguments: ZGrid, Hue_l, Sat_l, Hue_h, Sat_h/*/* ZGrid: the DEM !!!NOTE!!! The DEM must be FLOATING POINT/* Hue_l: Hue of the LOW elevation color/* Sat_l: Saturation of the LOW elevation color/* Hue_h: Hue of the HIGH elevation color/* Sat_h: Saturation of the HIGH elevation color/********************************/* BEGIN THE PROGRAM/********************************

207


/* Program Setup&args ZGrid Hue_l Sat_l Hue_h Sat_h&severity &error &routine bailout&term 9999display 9999/********************************GRID /* start GRID&describe %ZGrid%&sv cell = %grd$dx%/********************************&call v&call h&call s&call draw&call image&call exit&return/********************************/* ROUTINES/********************************&routine v&sv azim = 315&sv inclin = 45initial = hillshade(%ZGrid%,%azim%, %inclin%, shade)&describe initialxxg1 = (initial - [Value GRD$MEAN]) * 15 / [value GRD$STDV] + 95v = con(xxg1 <= 70, 70, xxg1 <= 99, int(xxg1), isnull(xxg1), 99, 99)kill (!xxg1 initial!) allv1 = con(isnull(v), 100, v)kill v allrename v1 vsetwindow v&return/********************************&routine h&describe %ZGrid%h = %hue_l% - (%hue_l% - %hue_h%) * ((%ZGrid% - %grd$zmin%)/(%grd$zmax% - %grd$zmin%))h1 = con(isnull(h), 0, h)kill hrename h1 h&return/********************************&routine s&describe %ZGrid%s = %sat_l% - (%sat_l% - %sat_h%) * ((%ZGrid% - %grd$zmin%)/(%grd$zmax% - %grd$zmin%))s1 = con(isnull(s), 0, s)kill srename s1 s&return/********************************&routine drawmape vgridcomposite hsv h s v&return/********************************&routine image

QUATERNARY LANDFORMS IN WISCONSIN—HOW HILLSHADING CAN BE USED

&sv nm = dem_hillr = hsv2red(h, s, v)g = hsv2green(h, s, v)b = hsv2blue(h, s, v)makestack stack1 list r g barc gridimage stack1 # %nm% tiffkill stack1 all

&return/********************************&routine exit&if [show program] ne ARC &then q&if [EXISTS h -GRID] &then

kill h all&if [EXISTS s -GRID] &then

kill s all&if [EXISTS v -GRID] &then

kill v all&return/********************************/* Perform Cleanup actions if Program Fails&routine bailout&severity &error &fail&call exit&return &error Bailing out of HSV.aml

209

Alaska Division of Geological and Geophysical Surveys Frank Ganley

Arizona Geological Survey Steve Richard

Arkansas Geological Commission Steven Hill Jennifer Perkins

Avenza Software Inc. Susan Muleme

California Division of Mines and Geology Dave Wagner

College of William and Mary Karen Berquist

Delaware Geological Survey Lillian Wang

Dynamic Graphics, Inc. Skip Pack

Environmental Systems Research Institute, Inc. Mike Price

Florida Geological Survey Jon Arthur Amy M. Graves

Geological Survey of Alabama Nick Tew

Geological Survey of Canada Eric Boisvert Boyan Brodaric Vic Dohar Dave Everett

Hewlett-Packard CompanyAl Berry

Idaho Geological SurveyTim Funderburg Loudon Stanford

Illinois State Geological Survey Curt Abert Sheena Beaverson Rob Krumm Barb Stiff

Intergraph CorporationJeff Hyatt

Iowa Geological Survey Mike Bounk

Kansas Geological Survey David Collins Elizabeth Crouse Gina Ross John Siceloff

Kentucky Geological Survey Warren H. Anderson Jason A. Patton Rick Sergeant Tom N. Sparks Xin-Yue Yang

Louisiana Geological Survey Robert Paulsell R. Hampton Peele

Maryland Geological Survey Lamere Hennessee

211

APPENDIX A

List of Attendees at theDigital Mapping Techniques ‘99 Workshop

[Grouped by affiliation]

Minnesota Geological Survey Tim Wahl

Missouri Division of Geology and Land SurveyHal Baker Chris Vierrether

Montana Bureau of Mines and Geology Debbie Smith Larry Smith Susan Smith

National Park Service Steve Fryer

Nevada Bureau of Mines and Geology Gary Johnson

New Jersey Geological Survey Zehdreh Allen-Lafayette

New Mexico Bureau of Mines and Mineral Resources Mark Mansell Dave McCraw Becky Titus

North Dakota Geological Survey Ann Fritz Karen Mitchell Ryan Waldkirch

Ohio Geological Survey Tom Berg Jim McDonald

Oklahoma Geological Survey Wayne Furr

Ontario Geological Survey Brian Berdusco

Oregon Department of Geology and Mineral Industries Mark Neuhaus

Pennsylvania Geological Survey Bill Kochanov Tom Whitfield

Tennessee Division of Geology Elaine Foust

U.S. Air Force Alyssa Perroy

U.S. Geological Survey Greg Allord Dave BedfordTodd Fitzgibbon Bruce Johnson Diane Lane Gary Latzke Bob Lemen Jonathan C. Matti Missy Packard John S. Pallister Peter Schweitzer Dave Soller Nancy Stamm Will Stettner Ron WahlGregory WalshJohn Watermolen

University of California, Santa BarbaraJordan Hastings

Utah Geological Survey Kent D. Brown

Virginia Division of Mineral Resources Rick Berquist Elizabeth Campbell Ryan Perroy

West Virginia Geological and Economic Survey Scott McColloch

Wisconsin Geological and Natural History Survey Bill Bristol Mike Czechanski Chip Hankley Mindy James Deborah Patterson Kathy Roushar


213

APPENDIX BWorkshop Web Site

215

APPENDIX C

List of Addresses, Telephone Numbers, and URLs for Software and Hardware Suppliers

[Information contained herein was provided mostly by the authors of the various articles and has notbeen checked by the editor for accuracy]

Adobe Illustrator, Photoshop, Acrobat, and PageMaker - Adobe Systems Inc., 345 Park Ave., San Jose, CA 95110-2704, (408) 536-6000, <http://www.adobe.com>.

ALACARTE – U. S. Geological Survey, <http://wrgis.wr.usgs.gov/docs/software/software.html>.

Anatech Scanner - Intergraph Corp. Corporate Headquarters, Huntsville, Al 35894-0001, (256) 730-2000,<http://www.intergraph.com>.

Apple Newton hand-held Personal Digital Assistant and Macintosh - Apple Computer Inc., 1 Infinite Loop, Cupertino,CA 95014, (408) 996-1010, <http://www.apple.com>.

Arc/Info, ArcView, ArcScan, ArcPress, ArcExplorer and ArcCad - Environmental Systems Research Institute (ESRI)Inc., 380 New York St., Redlands, CA 92373, (714) 793 2853, <http://www.esri.com>.

AutoCAD, AutoCAD Map 3.0, and MapGuide - Autodesk Inc., 20400 Stevens Creek Blvd., Cupertino, CA 95014-2217,(408) 517 1700, <http://www.autodesk.com>.

Canvas - Deneba Software, 7400 S.W. 87th Avenue, Miami, FL 33173, (305) 596-5644, <http://www.deneba.com>.

CorelDraw - Corel Corp., 567 East Timpanogos Parkway, Orem, UT 84097-6209, (801) 765 4010,<http://www.corel.com>.

Data General – Data General Corporation, 3400 Computer Drive, Westboro, MA 01580, (508) 898-5000,<http://www.dg.com/>.

DBASE database software - dBASE, Inc., 2548 Vestal Parkway E, Vestal, NY 13860, (888) dBA-SE32,<http://www.dbase2000.com/>.

Delphi - Inprise Corporation, 100 Enterprise Way, Scotts Valley, CA 95066, (831) 431-1000, <http://www.borland.com/>.

EarthVision - Dynamic Graphics Inc., 1015 Atlantic Avenue, Alameda, CA., 94501-1154, (510) 522-0700,<http://www.dgi.com>.

Genasys Products - Genasys Systems Pty Ltd, Level 13, 33 Berry Street, North Sydney, NSW 2060, Australia, telephone61-2-9926-2800, <www.genasys.com>.

GSMAP and GSMCAD - U.S. Geological Survey, <http://ncgmp.cr.usgs.gov/ncgmp/gsmcad/GSMCW5.HTM>.

GSView and Ghostscript - Aladdin Enterprises, 203 Santa Margarita Ave., Menlo Park, CA 94025, (650) 322-0103,email: [email protected], <http://www.cs.wisc.edu/~ghost/index.html>.

GTX-OSR vectorizing software - GTX Corporation, 2390 East Camelback Road, Ste. 410, Phoenix, AZ 85016, (800)879-8284, email: [email protected], <http://www.gtx.com/>.


Hewlett Packard DesignJet Series Plotters (various) - Hewlett-Packard Co 8000 Foothills Rd., Roseville, CA 95747, 1-800-PACKARD, <http://www.hp.com>.

IDRISI - Clark University, 950 Main Street, Worcester, MA, (508) 793-7526, <http://www.clarklabs.org>.

LtPlus raster-to-vector conversion software – <http://ftp.digital.com.au/pub/grass421>.

LT4X - AverStar, Inc., 4099 SE International Way, Portland, OR 97222, (503) 794-1344,<http://www.averstar.com/gis/software_products.html>.

Macromedia Freehand - Macromedia Inc., 600 Townsend St., San Francisco, CA 94103, (415) 252-2000.

MapInfo - MapInfo Corporate Headquarters, One Global View, Troy, NY 12180, (800) 327-8627,<http://www.mapinfo.com>.

MapPublisher - Avenza Software Inc., 3385 Harvester Rd. Suite 205, Burlington, Ontario, Canada L7N 3N2,<http://www.avenza.com>.

Micrografx Designer - Micrografx, Inc., Richardson, Texas, (972) 994-6525, e-mail: [email protected],<http://www.micrografx.com>

Microstation Products - Bentley Systems, Incorporated, 685 Stockton Drive , Exton, PA 19341-0678, (800) 236-8539.

Onyx Postershop - Onyx Graphics Corporation, Salt Lake City, UT, <http://www.onyxgfx.com>

ORACLE - Oracle, Inc., 500 Oracle Parkway, Redwood Shores, CA 94065, (800) ORACLE1,<http://www.oracle.com/corporate/sales_offices>.

Palm handheld organizers - Palm Computing, Inc., a 3Com Company, 5400 Bayfront Plaza, Mail Stop #10112, PO Box58007, Santa Clara, CA 95052-8007, (408) 326-5000, <http://www.palm.com>.

Pendragon Forms – Pendragon Software Corporation, P.O. Box 7350, Buffalo Grove, IL 60089, <http://www.pendrag-on-software.com>.

Petra software - geoPLUS Corporation, 8801 South Yale, Suite 380, Tulsa, OK 74137, (888) PETRA-65,<http:.//www.geoplus.com/html/petra.shtml>.

PLGR - Rockwell Collins Headquarters, 400 Collins Road NE, Cedar Rapids, IA 52498, (319) 295-5100,<http://www.collins.rockwell.com/government-systems/products/gpssmdex.shtml>.

Scitex - Scitex America Corp., 8 Oak Park Drive, Bedford, MA 01730, Fax: (781) 276-5709, <http://www.scitex.com>.

Smallworld database software - Smallworld Systems Inc., 5600 Greenwood Plaza Blvd., Englewood, CO 80111,(303)779-6980, <http://www.smallworld.co.uk/services/worldwide_contacts.asp>.

Sun, SUNOS and SOLARIS 2.5.1 - Sun Microsystems Inc., 901 San Antonio Rd., Palo Alto, CA 94303, (650) 960-1300,<http://access1.sun.com>.

Tektronix - Tektronix, Inc., 26600 SW Parkway, Wilsonville, OR 97070, (800)TEK-WIDE, <http://www.tek.com/>.

Trumatch - 50 East 72nd Street, Suite 15B, New York, NY 10021-4242, (800) 898-9100, <http://www.trumatch.com>.

Versatec plotters - Xerox Engineering Systems - Versatec Products, 5853 Ruserrari Avenue, San Jose, CA 95138 USA,(408) 229-3071,<http://www.engineeringsystems.com/index.htm>

Windows95 and WindowsNT - Microsoft Corp., One Microsoft Way, Redmond, WA 98052-6399, (425) 882-8080,<http://www.microsoft.com>.